Making Payments Boring, Making Disagreement Public: Odd Lots' First Conversation at Jackson Hole

The best rulers are those whose existence is merely known by the people… When the work is done, the people all say: “It happened naturally.”

—— Laozi, Tao Te Ching, Chapter 17 (Spring and Autumn period; author’s translation)

What This Episode Is About

On August 27, 2026, Bloomberg’s Odd Lots set up again in Jackson Hole, Wyoming, and their first guest was Kansas City Fed President Jeffrey Schmid — the man who throws the party. Tracy Alloway and Joe Weisenthal opened honestly: the backdrop is beautiful every year, but this year genuinely matters, because so many questions in the economy are unresolved.

This year’s symposium theme is financial innovation in payments, which sounds like an internal technocrats’ meeting. That’s exactly what makes the episode interesting. From a deliberately “boring” starting point, the conversation slides into the yield curve, why inflation is still in the threes, who the AI data center buildout is competing with, generational turnover in the labor force, and — a few weeks into Kevin Warsh’s chairmanship — what that “good family fight” inside the FOMC actually looks like.

What follows is my own synthesis after listening, not a transcript and not a substitute for anyone’s judgment. For the full conversation, go support the show directly.

Key Takeaways

1. The goal for a payments system is that nobody talks about it. Schmid puts it plainly: the Fed wants to make payments boring. Five to ten trillion dollars a day move through those pipes, and the measure of success is not praise — it’s the absence of attention. He later quotes Warsh approvingly: better for the Fed to be on page B12 than on A1.

2. Atomic settlement removes more than waiting time — it removes the buffer. For his whole career, he says, payments meant float and fees: friction and cost. Technology can now move money instantly and reconcile it instantly. But that has an unromantic consequence. If the payment is instant, there must be proven liquidity standing behind it to settle. So a conference about innovation lands on two very old words: duration and liquidity. A system that used to lean on elapsed time for flexibility now has to lean on stock.

3. He treats the long end as a supply-and-demand question, not a political one. With the 30-year above 5%, he sidesteps the usual “should the Fed control the long end” framing. His line: I’m not that smart, so I go back to supply and demand. If prices are moving, something is happening between supply and demand — true for bonds, true for corn and wheat. Right now the demand side for credit has a large new participant: the commercial sector borrowing for AI and data center buildout, competing with public-sector borrowing for the same pool.

4. Crowding out isn’t a theory to him — it’s what he hears every day. Asked whether firms in his district feel they’re competing with data centers for labor, equipment, or freight, his answer is: every day. Steel, copper, machinery — the same inputs pulled toward the same demand. He even notes grain futures surging. Hence his instruction to “peel the onion back”: figure out what share of the growth number belongs to this particular boom, because that bears directly on getting inflation back to two. He uses the word flywheel — the faster it spins, the more you need to know who it’s throwing off and where the risk is landing.

5. Do high yields do the Fed’s work? A conditional yes. Higher prices do change behavior; they can slow decisions. But he also allows the opposite: a business may see returns so high that 80 basis points on the ten-year is irrelevant to the model. The more useful distinction is which part of the curve governs which decision. The ten-year prices ten- and twenty-year commitments — whether to build the plant. The short rate prices working capital — whether to pre-buy inventory, whether another 25 to 50 basis points on the line of credit is worth it. Asked whether he’s ever heard of a factory cancelled because of a single 25 basis point hike, his answer is: very rare.

6. The stretch from three to two is hard for asymmetric reasons, not technical ones. He states it directly: the labor force is in a pretty good place, and on inflation “we haven’t done our job yet.” The closer you get to target, the harder it becomes, because everyone worries about overshoot — you’re either too slow or too aggressive. Recent meetings produced several dissents, which he calls thoughtful and consistent with the kind of debate the new chairman wants.

7. A dissent isn’t defiance — it’s a statement about weights. Asked what hurdle he has to clear before dissenting rather than merely voicing uncertainty, his framing is worth keeping: a dissent says I think the risks between the two mandates are weighted differently than you think they are. He describes himself as a transmitter — traveling the Tenth District, listening to what businesses and local leaders worry about, carrying it to the table, then carrying the committee’s thinking back home. He also admits dissenting was hard under the previous chair, because he deeply respected how that person thought.

8. The most human moment is about the labor force. Baby boomers are retiring at roughly four million a year, and he says he has signed more retirement letters in the last three months than in the prior three years. What worries him isn’t the headcount — it’s the intellectual muscle walking out the door. So he challenges his 25-, 35-, and 45-year-old economists and bankers with a question: can we use AI to transfer what the 65-year-old knows to what you need to know now, rather than waiting until you’re 55 or 65 yourself?

The hosts close on an observation: the new chairman’s task forces don’t appear to be in much direct dialogue with the regional presidents yet — Schmid suggests that comes once the research emerges. Joe’s summary is blunt: fresh eyes on data and communications are welcome, but the immediate question remains that inflation is still above 3%.

Going Further

Why the headline says “rate cut, bullish” and nothing you own moves

This is the first source of frustration for most people. The policy rate moves, the headline is emphatic, and the company you’re watching builds what it was going to build and shelves what it was going to shelve. You start to suspect you missed something.

This conversation offers a clean fix: first ask which part of the curve prices that decision. A plant, a data center campus, a production line — payback measured in decades — is anchored to the long end, which the policy rate does not directly govern. That’s why a 25 basis point hike almost never cancels a factory. It isn’t stubbornness; that particular bolt is fastened to a different plate. Whether to pre-buy inventory, whether to carry a heavier working capital balance, whether the extra 25 basis points on the revolver is worth it — those are short-end questions, and they respond fast.

Two practical habits follow. First, when a rate headline lands, separate capital-expenditure effects from working-capital effects: one is slow, one is quick, and blending them makes the market look irrational when it isn’t. Second, and easier to miss: the long end hits valuation and operations on two different clocks. The discount rate changes paper value today; orders and capex may not reflect anything until the next planning cycle. “Good news, no move” is often just that gap, not a lie in the news.

Keep his line: I’m not that smart, so I go back to supply and demand. When a price move has no obvious explanation, don’t reach for conspiracy or sentiment first. Ask who got bigger and who got smaller.

Is “AI is driving growth” good or bad for what I hold?

The buildout numbers set records weekly, you own a few things adjacent to it, and the instinct is to be pleased. What stopped me here was Schmid saying he hears about crowding out every day.

That one word settles something: the same demand carries opposite signs for different companies. For sellers of steel, copper, electrical equipment, and engineering services, it’s an order book. For anyone who must buy the same steel, the same freight capacity, the same scarce technicians to do something else, it’s cost inflation and longer lead times. If five names in your portfolio are all “AI-related,” ask each one whether it sits on the receiving or the paying side. And if all five sit on the same side, your diversification is cosmetic — you have placed one bet five times.

His instruction to peel the onion — separate the boom’s contribution from the rest of growth, because it determines the path back to two percent — is worth following all the way around, because it loops back to you. The boom lifts commodity and labor costs → inflation stays sticky → rates stay high or the long end drifts up → discount rates rise → the companies whose earnings sit furthest in the future get repriced hardest. That isn’t bearishness. It’s a warning that one variable may enter your model twice, with opposite signs.

Three questions worth answering per position: how much of this revenue growth belongs to this specific buildout cycle; what happens to the order book if the pace merely slows rather than stops; and the hardest one — does this business earn by selling shovels, or by finding gold? Shovel sellers feel best during the boom, and are the easiest to misread as structural when they’re cyclical.

Officials keep contradicting each other — who am I supposed to believe?

Follow Fed coverage for a while and the contradictions start to feel like noise. One says inflation isn’t beaten, another sees labor softening, the headlines collide inside a single week, and you conclude they don’t know either.

Schmid’s account of dissent turns that frustration into usable information. A dissent means: I weight the risks between the two mandates differently than you do. The disagreement is usually not about facts — it’s about ordering. They’re all looking at essentially the same data. What differs is who puts the long-run cost of inflation stuck in the threes ahead of the risk of a weakening labor market. Read that way, the clashing quotes stop being noise and become a distribution of weights you can count over time.

From there grows a more durable habit: follow the reaction function, not the tone. Hawkish or dovish tone can be rewritten by tomorrow’s print. The reaction function — what they will do when a given number arrives — is far more stable. Joe’s closing point identifies exactly this gap: the new communication style hasn’t settled, the market doesn’t yet have a firm handle on the reaction function, and until it does, every discussion of task forces and data reform runs into the same immovable fact — inflation is still above 3%.

There’s a bonus lesson that has nothing to do with markets. What Schmid says he admires isn’t the “family fight” metaphor itself but the posture behind it: being willing to hear people’s truths, because those truths may change how you think. He’s explicit that he dislikes the my-way-or-the-highway style. A committee of nineteen people, each backed by a serious team, that tolerates only one voice has wasted eighteen of them. That holds at any table where decisions get made.

Where to Look Next

• The Odd Lots episode itself, August 27, 2026 (there’s a video version; the hosts recommend it)
• The Kansas City Fed’s Jackson Hole Economic Policy Symposium, held each August, with papers and agenda published officially — next year is the 50th
• For “atomic settlement,” start with public central bank and BIS research on real-time gross settlement and tokenized settlement
• FOMC minutes and regional presidents’ published speeches remain the rawest material for observing who ranks which mandate first

The One Thing to Take Away

The spine of this episode is a single sentence: a system that removes friction has to prepare its buffer in advance.

Instant settlement sounds like it’s merely faster. But speed has a price — that waiting period was quietly acting as insurance. While money was in transit, the system had time to catch errors, move funds, plug gaps. Once settlement is instantaneous, that insurance disappears, and liquidity has to be proven up front. That is why a conference about innovation ends up talking about duration and liquidity. Every efficiency gain that removes a buffer is paid for; it just moves the cost from afterward to beforehand.

A practice for today: identify one thing in your life currently held together by a lag — it doesn’t have to involve money. Maybe the monthly bill clears only because the paycheck lands the day before. Maybe the report always gets finished the night before it’s due. Maybe the medication, the contact lenses, the pet food only get reordered on the day the last one runs out. Maybe it’s a relationship where the thing that needs saying keeps getting pushed to “next time we meet.”

Write that one thing down, then answer a single question: if that lag vanished tomorrow, how much would I need already on hand to be fine? Three days of cash, a draft written in advance, a spare box, or one sentence said tonight.

Then stock the first unit today. Not all of it — one. The value isn’t in whether that day ever comes. It’s in suddenly seeing how much of your life is being held up by “there’s still time.”

Raw Markdown

This article is an educational discussion of investment method. It is not advice to buy or sell any individual security, offers no target prices, and does not analyze any current holding. Investing carries risk; make your own decisions or consult a qualified professional.