Newsletter · 2026-08-31 · Ashutosh Agarwal

A Five Percent Long Bond Is Normal Again and the Dollar Absorbs the Strain - The Dollar Brief - Week of August 31, 2026

The Dollar Brief for the week of August 31, 2026. Podcast synthesis on why Jeff Snider calls a 5.31 percent 30-year yield barely a rounding error, Moody's decomposition putting at least half the move in term premium rather than inflation, former Fed president Jim Bullard asking why not hike in September, and Win Thin's point that a weak dollar and low yields cannot both be had.

The Dollar Brief

Week of August 31, 2026: A Five Percent Long Bond Is Normal Again and the Dollar Absorbs the Strain

For a week, the whole market stared at one number: the 30-year U.S. government bond yield touching 5.31%, its highest since 2007. The headlines wrote themselves: the bond vigilantes are back, foreigners are dumping America, the debt has finally broken the market. Treasury Secretary Scott Bessent got spooked enough to double the government's bond buybacks, which knocked yields down for a few hours before they crept right back up.

But step away from the flashing screens and the more interesting fight on the podcasts wasn't "how high do yields go." It was something more basic, and more important for anyone who holds dollars: is a 5% yield actually a crisis at all, or just a return to how things looked before a strange, cheap-money decade spoiled us? And underneath that: if the government is now willing to lean on the bond market to keep yields down, does the dollar have to be the thing that quietly pays the price?

That's what this issue walks through. One camp of very sober people says: calm down, this is normal. Another camp of equally sober people says: the size of the debt makes "normal" dangerous this time. A third group, including a former Federal Reserve official, says the Fed should stop dithering and actually raise rates. And the currency crowd is split three ways on where the dollar goes from here. Let's take them in turn.

(Quick vocabulary, used throughout. "FX" is just foreign exchange, where currencies trade. The "DXY," or dollar index, measures the dollar against a basket of big currencies. A "bond" is long-term government debt; a "T-bill" is short-term government debt that matures in under a year. A bond's "yield" is the interest it pays; prices and yields move in opposite directions, so when buyers back away and prices fall, yields rise. The "10-year" and "30-year" are the government's benchmark long-term bonds. "Bull steepening" is a specific shape-shift in the market: short-term yields fall faster than long-term yields, so the gap between them widens. "Term premium" is the extra yield investors demand as compensation for locking their money up for a long time; it can't be seen directly, only estimated. The "bond vigilantes" are investors who supposedly punish a government for reckless borrowing by refusing to lend except at painful rates. A "buyback" is the government buying back its own older bonds. "Yield curve control" is a central bank or treasury actively pinning yields at a chosen level. "Fiscal dominance" is when a government's debt load gets so heavy that monetary policy ends up bending to serve the budget rather than the economy. "Core PCE" is the Fed's preferred inflation gauge, stripping out food and energy. "Fed independence" means the Fed sets rates free of the elected government, the thing that keeps borrowing costs lower because markets trust it. "Debasement" is letting inflation quietly erode the real value of debt and savings.)

TL;DR

The contrarian case: this isn't a crisis, it's normalization. Jeff Snider argued the 30-year at 5.31% is "barely a rounding error," up only about 20 basis points in nearly three years, and that the falling 2-year yield proves this is ordinary "bull steepening," "not inflation or vigilantism" (Eurodollar University, Aug 19).
Even the man who coined "bond vigilantes" isn't panicking. Ed Yardeni called the move "orderly" and said "4% to 5% is kind of back to normal," adding he "really dislike[s] that phrase that interest rates are going to stay higher for longer" (Bloomberg Surveillance, Aug 19).
What's actually driving yields up: term premium, not inflation. Moody's Analytics decomposed the 10-year and found inflation expectations flat and "at least half" of the rise sitting in term premium, investors demanding more to hold the debt of a government running a ~6%-of-GDP deficit (Moody's Talks, Inside Economics, Aug 21).
The insider hawk: former Fed president says just hike. Jim Bullard noted core inflation has been projected above 3% for four years running, "that looks like a 3 percent inflation target," and asked, "why not go in September?" (Bloomberg Surveillance, Aug 19).
A September hike is closer than the base case suggests. TD's Gennadiy Goldberg said the bar is "incredibly low" and "we know three of them already want to hike… You don't need that many more people to basically overrule the Fed chair," and that a hike would actually pull long yields down (Bloomberg Surveillance, Aug 19).
The dollar's real trap: you can't have a weak dollar AND low yields. FX veteran Win Thin flagged the administration's core contradiction, wanting a cheap dollar to reshore manufacturing while also wanting low borrowing costs, as fundamentally "inconsistent," warning reserve status "works till it doesn't" (Bloomberg Surveillance, Aug 20).
The dollar-bear read: the currency absorbs the pressure. Saxo's John Hardy said the buyback is a "signal" of Treasury panic, and that once officials cap yields, "it's the currency that's going to absorb the pressure… very bearish to the dollar… very bullish hard assets" (Saxo Market Call, Aug 20).
The allocator's middle ground: range-bound dollar, but get defensive. Rebecca Patterson said she's "in kind of the range camp," no dollar collapse, while preferring "cash or short-term fixed income and… gold," and warning the U.S. bond market is starting to trade "a little bit" like a developing market's (Bloomberg Intelligence, Aug 22).
The reserve-currency warning. Brookings' Robin Brooks rated his worry "a 6 or a 7" out of 10, noting long yields have jumped "most of all in places that are highly indebted and/or politically dysfunctional… Japan, France, the UK, Italy, and of course also the United States," and that the market prices the 10-year yield ten years out at 6% (Marketplace, Aug 20).
The tell nobody should miss: the buyback wore off in hours. Across the board, the message was the same. The move was "financial engineering… shuffling the deck chairs" (Robin Brooks), a "shot across the bow" that was "small potatoes" (Win Thin). Yields were "already a third of the way back" before the session ended.

What's new

The contrarian who says everyone is reading the wrong number

Start with the loudest counter-take of the week, because it reframes everything else. On Eurodollar University (Aug 19), independent macro analyst Jeff Snider took direct aim at the crisis narrative. Yes, he acknowledged, the 30-year yield hit 5.31%, "its highest level since June 2007." But he says that headline is a magic trick. "A 5% long-term treasury yield sounds big and it looks big only if your chart begins, say, in December 2008." Zoom out: "During much of the 1990s, the 30-year treasury yielded between roughly 6% and 8%. In the 1980s, long-term yields spent years in double digits." His punchline: "when someone says the 30-year yield is its highest since 2007, what they're really saying is that it's returned to a level that was considered fairly low before an extraordinary era of disinflation, financial crises, and weak growth."

His strongest evidence is one most headlines ignore, the rest of the yield curve. If the market were truly bracing for runaway inflation or a buyers' strike, the whole curve should be selling off, especially the short end that tracks Fed policy. Instead, "while the 30-year has reached a new high, the two-year is going back lower again. That's not what a clean, economy-wide inflation repricing normally looks like. This is instead bull steepening," the ordinary process of a distorted curve regaining its normal upward slope as the market bets the Fed will eventually ease. "Normalization, not inflation or vigilantism."

And he's seen this movie before. "In late October 2023, the 30-year long bond went above 5% for the first time since 2007, and everyone told the same story then," deficits, vigilantes, the end of the low-rate era. "Then the yield fell, and fell sharply. By late December 2023, the 30-year yield had dropped by more than a full percentage point." Nearly three years later, he noted, the yield is "only about 20 basis points" above that 2023 peak. "20 basis points in nearly three years is not a bond market revolution. It is barely a rounding error." His closing line is the one to keep: "The loudest number in the bond market is not always the most important one… A market can make a new high without telling a new story." (Snider is a long-standing contrarian who thinks the "term premium" concept "isn't a real thing," a minority view, so treat his framework as a deliberate provocation. But his curve evidence is concrete and worth weighing.)

Remarkably, the man who invented the phrase "bond vigilantes" mostly agrees. On Bloomberg Surveillance (Aug 19), Ed Yardeni of Yardeni Research refused to hit the panic button: "It's been orderly. The levels are eye-catching. I get that. But it's been orderly." His line in the sand is 5%, and even there he shrugged: "I think 4% to 5% is kind of back to normal. I really dislike that phrase that interest rates are going to stay higher for longer because it implies that somehow or other we should be getting to lower interest rates." His one real complaint was aimed at the Fed's new chair, Kevin Warsh, who has punted big decisions to internal "task forces" due to report at year-end: Yardeni said Warsh "kind of cornered himself to do nothing until he hears from the task forces at the end of the year. Very bad idea."

If it's not inflation, what is pushing yields up? A clean autopsy

The most useful hour of the week for actually understanding the move came from Moody's Talks, Inside Economics (Aug 21), where the Moody's Analytics team did the unglamorous work of taking the 10-year yield apart. They split it into three pieces. Piece one, inflation expectations: "you get no explanation for the run-up in rates because inflation expectations are today right where they need to be," roughly the Fed's 2% target, and back to where they sat before this year's war-driven spike. Piece two, real short-term rates, "up 30, 35 basis points," which simply reflects the market pricing in "a couple rate increases" from the Fed. Piece three, the term premium, and that's where the action is. "It's really in the term premium… roughly speaking, it feels like at least half of the increase in long-term rates is the term premium."

So what's stuffed inside that term premium? Their answer was blunt: debt. "The first order problem is debt, just the rising U.S. debt burden." One of the economists, Martin, laid out the arithmetic: a deficit "close to 6% of GDP," a primary deficit (excluding interest) of 3% "when your economy is at full employment," publicly-held debt at 100% of GDP and, under benign assumptions, "120% 10 years from now." The mechanism is pure supply and demand: "the more debt there is, the more it's going to push up rates… I can shift that at the margins… but at some point, if there's more debt, the price of borrowing is going to increase. There's no way around that."

They also named the quieter fear lurking in the premium, the one that connects to gold and the dollar. Beyond plain default risk (which they dismissed as tiny for a top-rated borrower), there's "debasement risk": "Is there going to be a period in the future where… we're going to start to inflate away some of this debt? Were we trying to manipulate yields in order to keep that a little bit lower?" As one put it, "you can see this in the rise of the price of gold and similar assets over the past few months." On Bessent's buyback specifically, they were unimpressed. It's small (from $2 billion to $4 billion per operation, maybe $15 to $20 billion a quarter) and, worse, self-defeating as a signal: "what the Treasury really here is signaling is like we don't like where the yield is. So we might actually intervene to push that down… This is what we really call financial repression." Their odds of an actual bond-market blow-up, the 10-year climbing "well into the fives, maybe close to six" over the next 6 to 12 months, were sobering but not alarmist: one economist put it at 20% to 25%, another at 15%, a third at "10-15."

The other reason yields are up: the government now competes with the AI boom

There's a structural story stacked on top of the fiscal one, and MUFG's George Goncalves told it well on Bloomberg Surveillance (Aug 20). His framing: "sovereigns are competing for capital for the first time with the private sector. That's a big dynamic… The crowding out effect is now a real palpable thing. We're actually seeing crowding out between like AI issuance versus sovereigns." In plain terms, the tidal wave of debt that tech companies are issuing to build AI data centers is fighting the U.S. government for the same pool of savings, and both have to pay up. His silver lining for savers: "For the first time in many, many, many years, you are getting compensated… in the bond market again." And he read the one-day, 10-basis-point lurch in the 30-year as a positioning tell: "People were scared to actually be invested in the long-term sector. And they got a rude awakening."

The insiders and desks who want Warsh to stop dithering and just hike

Here the tone shifts from "stay calm" to "do something." The weightiest voice was a former Fed official. On Bloomberg Surveillance (Aug 19), Jim Bullard, the former president of the St. Louis Fed, made the credibility argument plainly. The committee keeps projecting core inflation above 3% at year-end, and has done so for four straight years. "That would be the same as in 2023, 2024, 2025, and now 2026. So that looks like a 3 percent inflation target," he said. "They have to take action to convince markets that they really want to get inflation back down to 2 percent." With growth strong ("GDP now for the third quarter… up at 4 percent"), unemployment low, and wealth at records fueling spending, "it's a good time to reestablish credibility on inflation fighting and go ahead and make a move." Asked directly whether the Fed should hike in September: "Well, I'd consider it. I don't think it's too bad of an idea… why not go in September?" On the July meeting minutes released that day, he predicted they'd "sound hawkish." (Bullard is no longer a voting official, but as a recent Fed insider his read on the committee's mindset carries more weight than a pundit's, worth separating from the newsletter-writer hawks of prior weeks.)

The practitioner echo came from TD Securities' Gennadiy Goldberg, who called the long-end sell-off "death by a thousand cuts." His firm's base case is a hold for the year, but he was candid about how fragile that is: "I think we're actually very, very close to a September hike, even though it's not our base case. I think the bar is actually incredibly low." The math of the committee, he noted, favors the hawks more than people assume: "we know three of them already want to hike. They dissented. You don't need that many more people to basically overrule the Fed chair." What would tip it? "One more payroll… one more CPI print," say 150 to 200K jobs, a 4% unemployment rate "for good reasons," and core PCE around 0.35% to 0.4%. And here's the counterintuitive kicker that both hawks lean on: a rate hike might lower long yields, not raise them. "I think the curve would flatten materially on a rate hiking cycle," Goldberg said, because a credible inflation-fighting Fed removes exactly the debasement premium the Moody's team described. He pegged 5% on the 10-year as "a good line in the sand" he doesn't expect to break.

There was even a subtle, under-the-radar development on the inflation data itself. As Marketplace (Aug 20) reported, the Bureau of Economic Analysis is changing how it calculates the PCE inflation gauge, tweaks that economists expect to make measured inflation look "a tenth of a percent to three-tenths of a percent lower" (it currently sits at 3.7%). UBS economist and former Fed staffer Alan Detmeister was careful not to cry foul, "that doesn't mean the BEA is cooking the books," but he raised an eyebrow at the timing: the change lands just as "you have an administration that has been talking about how they need to get inflation down," and "the lower inflation appears, the easier it is for the Fed to cut or at least not hike interest rates." A small thing, but it feeds the same worry running through the whole week, that the machinery around the dollar is being nudged, quietly, toward cheaper money.

The dollar's turn: three camps, one shared unease

All of the above lands on the currency. Here the week split cleanly into three views.

The trap (Win Thin). The sharpest structural point came from FX veteran Win Thin on Bloomberg Surveillance (Aug 20), who identified a contradiction at the heart of the administration's wish list. "The two things that are inconsistent are a weak dollar and lower yields," he said. Washington wants a cheaper dollar to make American factories more competitive and reshore manufacturing, "but if you're a foreign investor and you're holding US Treasuries, well, all of a sudden a weaker dollar is going to eat into your returns. You're going to have to ask for a higher return." Push the dollar down, in other words, and you push the yields you're trying to suppress up. He was measured on reserve status, "We get away with a lot because we are the face of the world's reserve currency, at least for now… it works till it doesn't. At some point, there's going to be a tipping point. This may be… a warning sign," and he expected more Treasury intervention, dismissing this first move as a "shot across the bow… small potatoes."

The bear (John Hardy). Saxo's John J. Hardy was the week's clearest dollar-down voice on Saxo Market Call (Aug 20). His logic: once officials signal they won't let long yields run free, foreign bondholders stop worrying about yield and start worrying about the currency, "so it's the currency that's going to absorb the pressure. That is indeed what we saw." His verdict: "This is very bearish to the dollar. It's very bullish hard assets." He noted the euro pushing up to $1.17 and the Swiss franc doing the heavy lifting (dollar-Swiss falling below 0.80), and flagged gold pressing against roughly $4,500, its 200-day average, with room toward "$5,000-plus" if it breaks through. His bigger thesis is a "fiscal dominance era" in which the government's debt load forces policy rates "far below inflation," and he openly wondered whether the eventual fix is to "redefine the inflation target to 3% instead of 2%." He was withering on the idea that Warsh will play tough guy: "He wasn't put into the Fed to be super tough on raising front end interest rates. I really don't believe that."

The middle (Rebecca Patterson). The allocator's view came from strategist Rebecca Patterson (formerly of Bridgewater and Bessemer Trust) on Bloomberg Intelligence (Aug 22). She's been wary of long bonds "for some time" and, faced with this market, "would prefer to be in cash or short-term fixed income and other types of diversifiers, dare I say it, gold," even favoring money-market-length duration over a 7-year bond, because "the yield pickup you have by owning a longer-term bond right now… is it worth the risk? And right now, I haven't thought it is." On the currency she planted a flag in the middle: "I'm more in kind of the range camp right now. So I don't think the dollar is about to fall off a cliff." But she offered the week's most quietly damning image of the U.S. bond market. Asked whether it's starting to look like a developing market's, she said, "A little bit, yes. I mean, if I wanted this kind of volatility, I'd go and invest in another country's bonds, not the U.S." Her list of what could force an "ugly" reckoning: an AI-stock-and-debt unwind that drags down "the entire stock market, the entire corporate bond market"; sticky inflation forcing the Fed to hike harder than expected; or contagion from a European debt crisis ("if France has a debt crisis, we will have contagion here").

The reserve-currency question, priced

Tying the dollar and the debt together, Brookings' Robin Brooks gave the week's cleanest reserve-status framing on Marketplace (Aug 20). His key observation is comparative: long-term yields "have risen massively, most of all in places that are highly indebted and/or politically dysfunctional. So places like Japan, France, the UK, Italy, and of course also the United States." The number that worries him isn't today's 10-year at 4.7%, "doesn't seem very worrying," but what the market implies for the future: "the 10-year yield 10 years forward… that's 6%." On a 1-to-10 scale, "I'm a 6 or a 7." His verdict on the buyback matched everyone else's: "I would call it financial engineering. It is really shuffling the deck chairs," because the actual problem is a 7%-of-GDP deficit that no buyback can fix. The most memorable line of the segment came from FHN Financial's Chris Lowe, on the Treasury's pivot toward short-term T-bill funding: "funding the U.S. government with T-bills is like buying a house with a credit card. The card company can reset the rate and does every time short-term interest rates change."

The debate

Is the bond sell-off a crisis or a return to normal? Mostly normal, but "normal" is more dangerous now than it used to be. The calm camp (Snider, Yardeni) has the better read on the market's mechanics: the falling 2-year yield, the contained 10-to-30 spread, and the fact that the identical panic in October 2023 fully reversed all argue that this is curve normalization, not a vigilante uprising. The worried camp (Moody's, Brooks, Patterson) has the better read on the stakes: at 60% debt-to-GDP a 5% yield is fine; at 100%-and-rising it strains the budget and invites the very intervention that spooks investors. Both can be right at once. The move is ordinary, the balance sheet it's happening on is not.

Should Warsh hike? The hawks have the stronger logical case; history and politics still favor a hold. Bullard and Goldberg make a genuinely counterintuitive argument, that a hike would lower long yields by killing the debasement premium, and note the committee already has three dissenters and a low bar. Against them: an election-year Fed rarely turns hawkish, the data has softened, and the new chair seems set on deferring to his own task forces. Watch the next jobs and inflation prints; Goldberg's tripwire (roughly 150 to 200K payrolls and 0.35% to 0.4% core PCE) is the thing that flips a hold into a hike.

Where does the dollar go? No collapse, but a real weakening bias, with a hawkish-Fed escape hatch. The bear (Hardy) and the range-camper (Patterson) actually agree on the mechanism: if the Treasury caps yields, the currency takes the strain, so the drift is softer. Win Thin sharpens why the government can't easily stop it, because a weak dollar and low yields are mutually exclusive goals. The one thing that would flip the dollar higher is the hawks' scenario: a credible, inflation-fighting hike would firm the currency even as it flattens the curve. So the dollar and the September Fed decision are the same trade.

The trades in play

Where podcasts named actual expressions this week (these are speakers' own views and positioning, not advice):

Up in quality, down in duration. Rebecca Patterson prefers "cash or short-term fixed income" and even money-market-length duration over longer bonds, judging the extra yield "not worth the risk" while rates are volatile (Bloomberg Intelligence, Aug 22). George Goncalves framed it as "dollar cost averaging in on every asset class, including the bond market," now that you're finally "getting compensated" (Bloomberg Surveillance, Aug 20).
Own gold as the debasement hedge. Both the bear and the worriers land here: John Hardy sees gold pressing $4,500 with room to "$5,000-plus" if it breaks (Saxo Market Call, Aug 20), and Patterson names gold as a core diversifier (Bloomberg Intelligence, Aug 22). The Moody's team tied gold's run directly to rising "debasement risk" (Moody's Talks, Inside Economics, Aug 21).
Short the dollar via the crosses, carefully. Hardy's dollar-bearishness showed up most in euro (up to $1.17) and especially the Swiss franc (dollar-Swiss below 0.80), though he flagged the franc move was partly a crowded-positioning unwind, not a clean new trend (Saxo Market Call, Aug 20).
The pain trade against all of the above: a September hike. If you believe Bullard and Goldberg, a hawkish surprise flattens the curve and firms the dollar, the mirror image of the consensus dollar-down, own-gold trade (Bloomberg Surveillance, Aug 19).

Read-throughs

The whole debate collapses into one date: the September Fed meeting. Nearly every thread, the dollar's direction, whether long yields break 5%, whether gold runs, hinges on whether Warsh hikes, holds, or (most likely, per the base cases) holds while sounding tough. The tripwire to watch is the next jobs report and core PCE print.
Watch the short end of the curve, not just the long end. Snider's most testable claim is that the falling 2-year yield disproves the inflation-panic story. If the 2-year starts climbing alongside the 30-year, the "it's just normalization" case breaks and the "regime change" case gets real.
The buyback is a signal, not a fix, so expect more of them. Everyone agreed the $2bn-to-$4bn move was too small to control anything and wore off within hours. The read-through: each time yields press the pain points (30-year above 5%, 10-year near 5%), expect another, larger intervention, and each one shifts a little more strain onto the currency.
The dollar is now the release valve. Win Thin's insight is the one to internalize: the government cannot get both a weak dollar and low yields. As long as it leans on yields, the currency is where the pressure escapes, which is why the bears and the range-campers both tilt the dollar softer.
The AI borrowing boom is now a macro rates story. Goncalves's "crowding out" point means the data-center capex wave isn't just an equity theme, it's a structural source of upward pressure on the government's own borrowing costs. A wobble in AI-related credit, as Patterson warned, could ricochet straight into Treasuries.

What changed this week

The narrative flipped from "crisis" to "orderly." After a week of 5%-yield headlines, the striking thing was how many serious voices, including the analyst who coined "bond vigilantes," described the move as normal and orderly rather than a break. The debate is no longer "is this a crisis" but "is normal now dangerous given the debt."
The hawkish case moved inside the tent. Prior weeks' calls for a rate hike came mostly from gold-newsletter pundits. This week it came from a former Fed president (Bullard) and a major sell-side desk (TD), with a concrete, low bar for September and a genuinely fresh argument, that a hike would pull long yields down.
The intervention lost its punch fast. The clearest new data point was behavioral: the Treasury's buyback knocked yields down for a few hours and they were "already a third of the way back" before the day was out. That failure-to-stick is what turns each future intervention into a bigger one, and pushes the adjustment onto the dollar.