Stagflation in the Present Tense
Why I believe it’s already here, and what I’d do about it.
There’s a word making the rounds in economic commentary right now, and nearly everyone who uses it takes care to keep it safely in the future tense. Economists debate whether we’re “headed for” stagflation. Analysts assign it probabilities. The word gets treated like a storm that might still turn out to sea, and the debate about its arrival has become a comfortable substitute for looking at what has already come ashore.
I’ll be the first to admit that I have a credibility problem on this subject. I run bearish by temperament, and by my own count I’ve predicted eight of the last two recessions. My own team applies a healthy discount to my gloomier pronouncements, and they’re right to do so. That’s exactly why I want to be clear about what I’m doing here: I’m not making a prediction. I’m reading numbers that have already been published.
Here’s what those numbers say. Headline inflation touched 4.2 percent this spring and still sits near 3.4 percent, well above target after months of grinding deceleration. Growth cooled to roughly 1.5 percent annualized in the second quarter. Brent crude has traded at $102, at $72, and back near $90 since February, and the fuel surcharge has worked its way into the price of everything that moves on a truck. And this week, in a move I’ll come back to, the United States Treasury started doubling the buybacks of its own long-term bonds. Slowing growth and rising prices, arriving together, courtesy of an oil shock: that’s the textbook definition of stagflation. We already have a mild case. The open question is duration, and duration is exactly the kind of question you want to face with your eyes open.
I learned a long time ago that problems don’t just disappear when we decline to look at them. In fact, they tend to grow in severity the longer they’re ignored. That’s doubly true for problems without an official referee. The committee that dates recessions typically rules about a year after the fact (which counts as prompt by the standards of official candor), and there is no committee for stagflation at all. Anyone waiting for somebody in Washington to make it official will receive the diagnosis as history.
What Makes Stagflation Different
To understand why I take even a mild case seriously, it helps to understand what makes stagflation unusual. In a normal economic cycle, growth and inflation move together. The economy runs hot and prices rise; the economy cools and prices ease. That pattern is what makes the Federal Reserve’s job possible, because its one real lever works on both problems at once: raise rates into the boom, cut them into the bust.
A supply shock breaks that machine. When something like an oil disruption hits from outside, it pushes prices up and activity down at the same time, and the single lever suddenly faces two fires. Raise rates to fight inflation and you deepen the slowdown. Cut rates to cushion the slowdown and you feed inflation. Every option is a trade-off between two mandates that now point in opposite directions, and the dilemma itself is the disease.
By that definition, the Fed is already living with it. Markets spent part of the spring handicapping a rate hike into a cooling economy, because oil kept threatening to reignite the price level. The 10-year Treasury sits near 4.7 percent, Tuesday’s sell-off pushed the 30-year to 5.34, and last week’s 30-year auction cleared at the highest yield since 2001, which tells you bond investors have stopped treating inflation risk as theoretical. You can imagine the mood in those committee meetings.
How We Got Here, and Why It Lingers
The trigger is no mystery. The American and Israeli campaign against Iran opened in the last days of February, and Iran answered by closing most of the Strait of Hormuz, the chokepoint that carries roughly a fifth of the world’s oil. An energy shock behaves like a tax with perfect coverage: it raises the cost of producing, shipping, and driving while draining the discretionary spending that would otherwise become someone else’s revenue. Brent spiked toward $102 in the spring escalation, and American gasoline brushed $4.05 a gallon at the worst of it. Prices fell to $72 in late June when a memorandum of understanding promised a path out, then climbed back through the high 80s once the memorandum’s 60-day window expired with nothing durable behind it. BNP Paribas modeled the shock at roughly 0.4 points off advanced-economy growth and 1.1 points added to inflation, and that’s their central scenario, the one that assumes the conflict cools from here.
I wouldn’t count on it cooling, and an old Cold War strategist explains why. Thomas Schelling drew a famous distinction in Arms and Influence between deterrence and compellence. Deterrence asks your adversary to keep doing nothing, and it can wait forever. Compellence demands a visible act of compliance by a deadline, and it has to stay in motion until it gets one. Schelling’s rule was that it is far easier to deter than to compel, and Hormuz is now a compellence problem for both sides. Tehran wants sanctions relief and reparations before the strait reopens. Washington wants the strait open before it discusses anything else. When two parties each insist that the other move first, you get exactly what the oil market has priced all summer: a long, expensive grind.
Small-business owners have been riding that grind in real time. The NFIB optimism index fell to 95.8 in March, its low for the cycle, during the exact weeks when owners reported absorbing fuel and input costs they couldn’t fully pass along. By July the index had rebounded to 99.8, its best reading in nearly a year, as oil calmed down. I would love to read that recovery as resilience. Unfortunately, I’ve been around long enough to suspect it might also be what Galbraith called the extreme brevity of the financial memory. The truth is I can’t tell which one it is, and that uncertainty bothers me more than the March number did.
The Case for Calm
Plenty of smart people think the worriers are getting ahead of themselves, and their case deserves a full airing. Eugenio Aleman at Raymond James puts the odds of a genuine stagflationary episode at “very low.” Gregory Daco at EY argues that everything turns on duration, and that the baseline outcome is a temporary inflation bump. The Council on Foreign Relations concluded this spring that a repeat of the 1965-to-1982 era is unlikely, and the structural arguments hold up well. America now produces and exports the oil it once imported, which cushions the blow and hands parts of Texas and North Dakota a boom inside the squeeze. The wage-indexation clauses that hard-wired 1970s pay to 1970s prices are mostly gone. The Fed has four decades of inflation-fighting credibility that Arthur Burns never enjoyed. The economy runs on services and software, unemployment sits near 4.2 percent, and the AI investment boom, whatever its excesses, is a productivity tailwind with no 1974 equivalent. On the history, the calm camp wins, and I’m happy to concede the point. I don’t expect gas lines, price controls, or double-digit inflation prints.
My problem with the comfortable conclusion is simple: it assumes the damage has to arrive the way it arrived last time, through the price level. We’ve spent the past fifty years building a different set of vulnerabilities, and this week the biggest one stepped into plain view.
When Plumbing Becomes Policy
This week, the Treasury Department announced that it will at least double its long-term buyback operations, from a $2 billion cap to a minimum of $4 billion per operation across the 10-to-30-year sectors, beginning September 9. The official language calls it “liquidity support,” and the official rationale points to the “significant volume of high-quality offers” Treasury keeps receiving in the long end, which is a polite way of saying that a great many people would like to hand their long bonds back to the government. The announcement came the day after that sell-off took the 30-year to 5.34 percent, and a week after the auction that cleared at 2001 levels. Yields fell on the news, stocks cheered, and gold jumped, because everyone worked out the translation at the same time.
Here’s the translation. For a decade, Treasury buybacks were boring liquidity plumbing, a few billion dollars of housekeeping that kept older bonds tradable. Conducted at the highest long-term yields in a generation, the day after a sell-off, they become something else entirely: the world’s largest borrower supporting the price of its own debt because private demand keeps asking for a better deal. Some of that missing demand has a surprising address. The AI buildout is being financed with hundreds of billions of dollars in new corporate bonds, and every buyer who reaches for that richer yield is a buyer who didn’t show up at the Treasury auction. Add war-driven inflation fears and a federal debt load that grows in every season, and the long end of the curve has become the one market Washington can least afford to leave alone.
I understand the temptation, and I want to be plain about where it leads. The government funds these purchases by issuing even more short-term debt, which means it’s swapping long promises for short ones at the exact moment inflation makes short promises expensive to keep. Easing financial conditions while consumer prices run a full point above target adds fuel to the inflationary fire, and that’s worse than illogical: it’s a catastrophic mistake, and the long end is where it will implode. A support operation announces the absence of real demand more loudly than any failed auction could, and the investors who hear that announcement respond by demanding more compensation, which invites a bigger operation, which sends a louder signal. We’ve run this experiment before. The Fed pegged long-term yields through the 1940s while inflation ran into double digits, and it took the Treasury-Fed Accord of 1951 to shut the arrangement down before it consumed the central bank’s credibility. The peg always looks free at the start. It never is.
Don’t Ignore Obvious Problems
A decade ago, in my first book, I wrote a chapter called “Don’t Ignore Obvious Problems.” The case study was Greece, where the math simply didn’t work, and where everyone in charge preferred to kick the can down the road until the problem grew beyond managing. I’ve been thinking about that chapter a lot lately, because the obvious problem in front of us today is sitting on the balance sheets of America’s regional and community banks.
The numbers deserve to be read slowly. Commercial real estate makes up 44 to 48 percent of loans at regional banks, against 13 to 19 percent at the money-center giants. The FDIC’s 2026 Risk Review puts the median CRE concentration at banks between $1 billion and $100 billion around 300 percent of Tier 1 capital, the level that triggers heightened supervisory attention, and hundreds of community institutions live at or above that line as a normal feature of their business model. Somewhere between $1.5 and $2 trillion of commercial real estate debt matures through the end of 2026, and every one of those loans arrives at the same fork: refinance at today’s rates or sell the building at whatever price the new math supports. Neither branch is comfortable for the lender. Klaros Group screened roughly 4,000 banks and found 282 carrying both heavy CRE books and deep unrealized losses on their securities portfolios, and the office sector shows what trouble looks like when it arrives: Manhattan office delinquencies rose more than 1,000 percent in the twelve months through January 2024, and the vacancies behind that number have proven stubbornly durable. Notice, too, that those unrealized securities losses are marked against exactly the long-dated bonds the Treasury is now propping up. If the long end gives way the way I fear it might, the refinancing wall and the bond losses arrive together.
Two old economists explain why I watch this so closely. Hyman Minsky taught that long stretches of calm quietly move borrowers from loans their income can retire toward loans that survive only by being refinanced; stability, he argued, reprices recklessness as prudence, one renewal at a time. Charles Kindleberger, whose Manias, Panics, and Crashes tracks four centuries of these episodes, added the timing: stock manias and property manias inflate together, but they deflate on different clocks. Equities crash fast and in public. Property declines slowly, over years, because the debt is long-dated and nobody can call it overnight. A refinancing calendar is the slowest clock in finance, and ours happens to strike between now and the end of next year.
This is where stagflation stops being an abstraction and becomes a transmission mechanism. The inflation half keeps rates high, which raises banks’ funding costs and their borrowers’ refinancing burden in the same motion. The stagnation half softens the tenants, the rents, and the local businesses that keep their deposits at those same banks. A stressed bank does the rational thing and tightens, the withheld credit comes straight out of the working capital of small businesses already paying the fuel surcharge, and the resulting slowdown circles back into the bank’s own loan book. I watch this loop from close range. At B:Side, our SBA lending runs through exactly these community banks and into exactly these businesses, and I’ve learned that renewal conversations tell you more than any index does. When bankers start slowing down, we feel it months before it prints.
Who Actually Pays
A stagflationary economy is an averaging machine, and the averages flatter it. Energy producers are booming. Large firms hedge their fuel, ladder their debt, and pass costs along. Roughly half of American consumer spending now comes from the top tenth of households, which props up the aggregate numbers while the median family absorbs the shock. Energy and food take a far bigger share of a small paycheck than a large one, real wages have lagged the sticky part of inflation, and layoff risk concentrates in the energy-sensitive sectors least able to carry idle payroll. I see the same split from the front of a classroom. My students read about 4 percent unemployment and a record stock market, then go home and watch their parents cut back. The average is doing fine, the median is tired, and that gap, more than any single statistic, is what stagflation actually means.
What I’d Do Right Now
If you own a business, start with pricing. March taught us that absorbing input costs is a strategy with a deadline, so reprice deliberately and early, and explain the increase to your customers like the adults they are. Next, measure your energy exposure per unit of whatever you sell, because the strait can close again faster than you can rewrite a contract. If you have a loan maturing within the next eighteen months, call your banker this month, while it’s still a conversation rather than a deadline, and fix every rate you can defend; if Washington’s buybacks hold the long end down for a few months while you do it, accept the gift. Get to know your bank the way it knows you: its concentration, its appetite, who actually holds your note. Finally, hold more cash than the efficiency textbooks recommend. Cash buys time, and time is the scarcest asset in a tightening.
If you lead a team, tell them the truth about the year. Your people are living the median experience while the headlines describe the average one, and that gap erodes trust faster than any bad news will. Budget for a range on energy prices instead of a point estimate, and watch the indicators that will actually settle this: credit spreads, lending standards, renewal terms, and the size of the Treasury’s next buyback. Most of them are dull. All of them are decisive.
The word itself carries a useful lesson. “Stagflation” was coined in 1965 by a British politician named Iain Macleod, who looked at slowing growth and rising prices and told Parliament his country faced “the worst of both worlds.” The name arrived before the fight did, and that’s the right order, because nobody can fight a condition they refuse to name. Our version is milder than his, at least so far, but it’s here. If I’m wrong, and the strait reopens, prices settle, and the buybacks fade back into obscurity, you’ll be left holding extra cash, longer-dated debt, and a banker who knows your name; there are worse fates. If I’m right, those same preparations will be the difference between a hard year and a dangerous one. Either way the moves are identical, and the time to make them is now. After all, problems don’t disappear just because we decline to name them. More often than not, they grow. Things are going to get interesting, that’s for sure.