The Second Signal

Date: 2026-06-20

Author: Wealth & Means Staff

Source: https://wealthandmeans.com/essay/the-second-signal-wealth-and-means-episode-37

The War on Poverty debate, Warsh's Fed debut, Ozempic's exercise paradox, PFAS UV light breakthrough — the week's real signals, decoded.

TL;DR

The thread underneath this episode is the same one it always is: what you see, and what you don't. Kevin Warsh held rates and used forty-three minutes to tell you the era of 'watching carefully' is over. Anthropic and OpenAI filed S-1s within days of each other — but the race isn't about valuation, it's about who gets to set the precedent. Ozempic users are losing weight and moving less, not more — and the reason is physiology, not laziness. UV light can destroy forever chemicals using nothing but the water they're already in. A Swedish lab extended the useful life of superconductors by changing what they sit on. And the commodity price of coffee crashed while the grocery price hit a forty-six-year record. The signal was always there. The question is whether you were looking at the right surface.

Key Takeaways

The thread underneath this episode is the same one it always is: what you see, and what you don't. This week, a new Fed chair spent forty-three minutes telling you the era of "watching carefully" is over. Ozempic users are losing weight and sitting more — not less — and the reason is physiology. UV light can destroy forever chemicals using nothing but the water they're already in. A Swedish lab extended the practical life of superconductors by changing what they sit on, not the superconductor itself. And the commodity price of coffee crashed forty percent while grocery prices hit a forty-six-year record.

The pattern: the signal was always there. The question is whether you were looking at the right surface.

What You Didn't See in the News

Kevin Warsh sat down at the Federal Reserve's podium for the first time on Wednesday, and spent forty-three minutes telling you exactly what kind of Fed chair he plans to be. The headline — rates held at three-and-a-half to three-and-three-quarters percent, for the fourth meeting in a row — wasn't the story. The story was everything that surrounded it.

Warsh is the seventeenth Fed chair in the institution's history. He was confirmed by the Senate in May by a vote of fifty-four to forty-five — the most divisive confirmation in Federal Reserve history. He was forty-five years old when he was sworn in. And for the past fifteen years, he's been publicly on record saying, in almost exactly these words: "inflation is a choice." At his first press conference, he said it again. He wants a narrower Fed. Less communication, more discipline, less guidance about what it might do and more focus on what it's actually doing.

The market saw this as hawkish. The committee's projections now raise the odds of a rate hike later this year — not a cut. That's a meaningful pivot from where we were ninety days ago, when one or two cuts by December were still the market consensus. Warsh is essentially telling you the era of "we're watching carefully" is over. In its place: the era of "we told you what we're watching for."

For investors, that shift has downstream consequences. Duration risk re-prices when the Fed gets more predictably hawkish. And if Warsh actually delivers on the institutional reform he's signaled — less Fed speak, fewer forward guidance signals — every market participant who built a strategy around interpreting the Fed's tone is going to have to recalibrate.

The reason this surfaced with such clarity this week: it was his first meeting. Not a test run — an actual debut. And he passed the audition for the people who hired him, while unsettling the people who weren't expecting a pivot this fast.

Jerome Powell spent four years threading needles. Warsh just picked up the needle and said he'd rather use scissors.

The IPO window that SpaceX blew open last week has two more mega-tests ahead — and this week, the race between them got considerably more concrete. Anthropic filed a confidential S-1 with the SEC on June first, targeting a valuation of approximately nine hundred and sixty-five billion dollars. To understand what that number is sitting on: Anthropic's annualized revenue was nine billion dollars in January. By May, it was running at forty-seven billion. That's a five-times increase in five months. OpenAI filed its own confidential S-1 days after Anthropic. Goldman Sachs, JPMorgan, and Morgan Stanley are split across the two underwriting rosters. The consensus timing for both offerings is Q4 2026, with Anthropic targeting October.

The sequencing matters because of what it does to valuation precedent. The company that goes first sets the revenue multiple against which the second gets measured. Anthropic is racing to establish that framework before OpenAI can; OpenAI would prefer to be the benchmark rather than be evaluated against one someone else set. For the investment banks, this is a scheduling problem as much as a financial one — the two largest tech IPOs in history can't land in the same two-week window without cannibalizing demand. Someone has to blink first.

What makes the sequencing consequential beyond the two companies: the combined anticipated raise for these two offerings exceeds one hundred and sixty billion dollars. SpaceX raised seventy-five billion and was the largest IPO in history. The million retail investors who bought SPCX in the first hours of trading are now making a decision — hold a company at a two-trillion-dollar valuation, or sell and wait for the next two. That behavior, aggregated across a million accounts, is the leading indicator for how the AI company IPO window actually prices.

Two companies racing to set the valuation ceiling the other will be judged against. Only one gets to go first.

That same pattern — something not working the way the optics suggested it should — shows up in tariff data that came into sharper focus this week. The conventional narrative around tariffs is that they protect domestic producers and redirect demand toward American-made goods. The data tells a more complicated story.

Research tracking retail pricing since October 2024 has found that the cheapest product varieties — the store brands, the value tiers, the bottom-of-shelf options — saw prices rise an average of five percent. Premium equivalents rose about two and a half percent over the same period. The average American household is now paying approximately fifteen hundred dollars more per year as a direct result of the current tariff regime, which the Yale Budget Lab has called the largest tax increase as a percent of GDP since 1993.

The mechanism isn't difficult to understand once you see it. Luxury goods are often domestically manufactured or sourced from countries with different tariff rates. Value goods are overwhelmingly imported from China and Southeast Asia — which is precisely the supply chain the tariffs were designed to hit. So the tariff pass-through is heaviest exactly where lower-income households shop. Furnishings, clothing, and electronics are the hardest-hit categories. Clothing is particularly regressive because lower-income households spend a higher share of their income on apparel than higher-income households do — so the same price increase lands harder.

The Dallas Fed published analysis this week showing that tariff effects on PCE prices peaked in the first quarter of 2026 and are now flowing into real-time retail data. The story isn't new — but the confirmation that it's embedded in the price level now is.

A tax that costs more when you have less isn't trade policy. It's just policy.

Another story gaining momentum this week: people taking GLP-1 medications are losing weight — and moving significantly less while they do it. A study presented at ENDO 2026, the Endocrine Society's annual meeting in Chicago, analyzed Fitbit data from seven hundred and fifty-three adults with obesity who were prescribed GLP-1 drugs like semaglutide, dulaglutide, or tirzepatide — the class that includes Ozempic, Wegovy, and Mounjaro. The finding: average daily steps dropped from five thousand and forty-seven to four thousand four hundred and eighty-seven after starting the medications. That's a loss of five hundred and sixty steps per day. Moderate-to-vigorous activity dropped by nearly six minutes daily. The declines were steepest among men and among people with musculoskeletal pain.

The hypothesis going in was intuitive — almost assumed. You lose weight. You feel better. You move more. The feedback loop seemed obvious. But what appears to be happening is closer to the opposite: the appetite suppression is so effective that the body is consuming dramatically fewer calories, and the biological response is to conserve energy by reducing spontaneous movement. That's not laziness. That's physiology.

The clinical stakes are real. GLP-1 drugs don't just reduce fat mass. They reduce lean muscle mass too. When you simultaneously reduce caloric intake through appetite suppression and reduce physical activity through this energy-conservation response, you risk losing the very muscle and bone density that give weight loss its long-term health dividend. The researchers aren't saying don't use these drugs. They're saying: the drug without a structured exercise protocol may be delivering a narrower form of health improvement than we thought.

This got significant attention this week because the Fitbit data gave it unusual credibility — it measured actual behavior, not self-reported activity, and the sample size was large enough to carry statistical weight.

The drug works exactly as intended, and that's the problem.

That same pattern — finding what was hiding in the chemistry — surfaces in a very different context. Scientists published research this week revealing that intense ultraviolet light can destroy PFAS compounds — the so-called "forever chemicals" found in everything from non-stick pans to municipal water supplies — without any added chemical reagents.

Here's how it works. When water is exposed to high-energy UV light, specifically at wavelengths below three hundred nanometers, the energy generates hydrogen radicals from the water molecules themselves. Those radicals are highly reactive, and when they encounter PFAS compounds, they attack the molecular bonds that make PFAS so persistent. In a five-hour laboratory test, one key PFAS compound called GenX degraded by forty-nine percent. That's not full destruction — but it's meaningful, and it's happening without any chemical input beyond light and water.

The researchers at the University of New South Wales, where this work originated, are careful to note that scaling this from a lab environment to a municipal water treatment plant is a significant engineering leap. But the underlying mechanism is now understood in a way it wasn't before. And that distinction matters: if you know exactly what's driving the degradation, you can design systems to maximize it.

The stakes are considerable. PFAS contamination affects an estimated two hundred million Americans according to the Environmental Working Group. Cleanup efforts have been chronically underfunded partly because effective treatment methods have been prohibitively expensive and technically complex. A scalable approach that requires nothing but a UV lamp and contaminated water would fundamentally shift the economics of remediation — from "how do we manage this problem" to "how quickly can we reverse it."

Two hundred million people drinking contaminated water, and the solution was in the water the whole time.

Which brings us to another materials science story getting attention this week — this one out of Sweden, and with implications that stretch from MRI machines to quantum computers. Researchers at Chalmers University of Technology in Gothenburg published findings showing that by subtly sculpting the surface beneath an ultrathin superconducting material — changing the texture of the substrate it sits on — they could make that material maintain its superconducting state at higher temperatures and under much stronger magnetic fields.

To understand what's at stake here: superconductivity is the state in which a material conducts electricity with zero resistance. For most materials, it only occurs near absolute zero — well below minus two hundred Celsius. The holy grail of materials science for decades has been finding materials that superconduct at higher temperatures, ideally at room temperature or ambient conditions. Earlier this year, a team at the Texas Center for Superconductivity broke the record for the highest superconducting temperature at ambient pressure — one hundred and fifty-one Kelvin. Still extremely cold by human standards, but a thirty-year record.

The Chalmers work isn't a temperature record. It's a durability record. A superconductor that can hold its state under physical and magnetic stress is a superconductor you can actually build something around. MRI machines, maglev trains, power transmission grids, and quantum computers all depend on superconducting materials — and the cost and fragility of maintaining those materials under operating conditions is one of the main reasons these technologies haven't scaled faster.

The paper is already circulating in quantum computing and semiconductor research communities where substrate engineering is a live commercial problem.

We've been trying to make the superconductor stronger. Turns out, the fix was what it was sitting on.

Moving to trade — and a confrontation that's been building quietly for months. The European Union is moving toward the most consequential trade action against China since the steel tariffs of 2018, and the country with the most to lose isn't France or Italy. It's Germany.

European Commission President Ursula von der Leyen is expected this week to outline new EU trade defense tools specifically targeting low-cost Chinese goods entering European markets. The immediate focus is electric vehicles. EU officials have been investigating whether Chinese EV manufacturers — BYD, SAIC, and Geely among them — are pricing their vehicles in ways that reflect state subsidies rather than market economics. Beijing has already responded: if the EU implements new mechanisms, China will take retaliatory measures.

Here's why Germany is caught in the middle in a way that's structurally uncomfortable. BMW, Volkswagen, and Mercedes generate between twenty-five and thirty-five percent of their revenues in China. Beijing's preferred retaliatory tool isn't a blanket tariff on European goods — it's targeted measures against the sectors most politically exposed to the relationship. A thirty percent Chinese import tariff on European luxury vehicles would hit German auto harder and faster than any broad trade restriction. That gives Berlin a direct financial incentive to lobby Brussels toward restraint — which is exactly what's been happening in internal EU negotiations for months.

Germany built the best cars in the world. Now those cars are also Beijing's most useful lever.

And in the Western Hemisphere — on Sunday, Colombia holds its presidential runoff. The first round produced a tighter-than-expected result: right-wing political outsider Abelardo de la Espriella at forty-four percent, against left-wing senator Iván Cepeda at forty-one percent. The gap has narrowed further in polling — some surveys show it within two to three points.

The policy contrast is as sharp as it looks. De la Espriella wants to loosen investment restrictions, reduce regulation on hydrocarbons, and rebuild Colombia's fossil fuel export infrastructure. Cepeda wants a renewable energy transition, land reform, and a very different relationship with foreign capital. And underneath both platforms is a China dynamic that doesn't get enough attention in English-language coverage. Chinese mining and infrastructure investment in Colombia has been growing steadily. De la Espriella's platform would likely accelerate that investment. Cepeda's would potentially redirect or slow it.

For anyone with Latin American commodity or infrastructure exposure, Sunday's margin matters more than Sunday's winner.

A wide de la Espriella victory reinforces the existing capital flow thesis. A Cepeda upset — which polling doesn't rule out — forces a significant repricing of those flows.

In Colombia, the question isn't just who runs the country. It's who gets to run the mine.

Speaking of gaps between projected outcomes and reality — the FIFA World Cup is two weeks into its group stage, and the US hospitality industry is calling it a non-event. Eighty percent of US hotels surveyed in World Cup host cities reported bookings below projections. In Kansas City, eighty-five to ninety percent of hotels said the same thing. Atlanta saw hotel bookings drop by eighty-three thousand rooms compared to the same period last summer.

There are multiple factors at play. FIFA delayed ticket distribution late enough that fans couldn't finalize travel plans before the booking windows for affordable accommodations closed. European bookings to US host cities are down fourteen percent. There's also an accommodation substitution story: when hotels spike prices around mega-events, a meaningful cohort routes around them entirely and books short-term rentals. The spend still happens — it just doesn't show up in hotel occupancy data.

The underlying tension here is structural. FIFA projects eighty billion dollars in gross economic output from this tournament. The organization itself is expected to pocket around eleven billion dollars in net profit. What remains for host communities — the cities that spent years and billions upgrading infrastructure — appears to be substantially less than what was modeled. That gap between what FIFA promises and what host cities receive is a recurring pattern.

The money machine works. It's just that FIFA is the machine.

Let's stay with things that aren't performing as the models predicted — and this one hits closer to home. A new international study published this week found that middle-aged Americans are lonelier, more depressed, and showing steeper cognitive decline than comparable generations studied before — and worse than their own parents at the same age.

The research compared middle-aged adults across several countries — the US, Canada, the UK, Germany, Australia. The US results were outliers. Americans in their forties and fifties are reporting worse memory, higher rates of depression, and more social isolation than the same cohort studied a decade ago, and worse than comparable cohorts in peer nations measured in the same period. The researchers controlled for income and education levels, which means the divergence isn't simply explained by economic stress.

What they didn't fully isolate — and this is where honest science acknowledges its limits — are the structural factors that might be driving the US-specific divergence: smartphone adoption curves, the specific character of American work culture, the erosion of civic and religious institutions, the social fragmentation that accelerated after 2020. Canada, the UK, Germany, and Australia all underwent similar technological and economic pressures. They don't show the same pattern. That makes this an American phenomenon, not a universal generational one.

The second-order consequences are significant and slow-moving. Cognitive decline and social isolation in midlife are strong predictors of dementia risk, cardiovascular outcomes, and workforce productivity. This is an early signal in a long curve.

The loneliest generation in American history didn't get there alone.

Moving to attention markets — the creator economy crossed two hundred and fifty billion dollars this year, growing at nearly twenty-three percent compounded annually. But the more interesting number is the one underneath it.

Social media creator revenue is up sixteen percent in 2026 to twenty point six billion dollars. Micro- and nano-influencers now account for forty-five and a half percent of all influencer marketing spending. And a quiet but meaningful trend is pulling capital away from the big platforms entirely: the rise of the platform-agnostic creator. These are creators whose primary income streams don't run through any single platform's algorithm or payment system. Email lists. Private communities. Subscription layers. Direct commerce integrations. Revenue that doesn't show up in the platform dashboards that analysts report from.

The logic behind this shift isn't complicated once you see it. TikTok's regulatory volatility showed creators what it looked like to build a business on uncertain ground. YouTube's ad policy shifts have done the same thing at intervals. The creators who've been at this long enough learned that platform dependency is a liability, not an asset.

That last point matters for anyone tracking digital advertising markets. The two hundred and fifty billion dollar headline figure probably understates the actual economic activity, because the platform-agnostic revenue is genuinely difficult to capture in industry data.

The most durable businesses always end up owning the relationship. The platforms forgot that creators could figure that out too.

Here's a number worth sitting with: Arabica coffee futures are down roughly forty percent from their 2025 peak. And the average bag of roasted ground coffee at a US grocery store just hit nine dollars and seventy-two cents — the highest price on record since the Bureau of Labor Statistics started tracking it in 1980.

The commodity crashed. The retail price went the other way. Both of those things are true at the same time, and the gap between them is worth understanding.

Brazil's 2026/27 harvest is tracking at somewhere between seventy and seventy-five million bags — up twenty percent from last year, a potential all-time record. StoneX projects the global coffee surplus for 2026 will reach ten million bags, the biggest surplus in six years. Arabica futures have already priced in most of that, falling from above four dollars a pound to around two sixty-five. The commodity market moved.

The retail market has not. What you're watching is a textbook version of "rockets and feathers" pricing — commodity costs go up like a rocket and come down like a feather. Roasters signed forward contracts when prices were high and are working through that inventory now. Retailers negotiate seasonal pricing, not real-time pricing.

If the surplus holds and the commodity price stays where it is, meaningful retail price relief is coming — probably in the second half of 2026. The commodity market has already moved. The consumer market is running three to six months behind it. That delay is visible in the futures curve and invisible at the register.

Wake Up Ready

Here's your macro weather report for the week ahead.

Start Sunday, because the clock is already running. Colombia holds its presidential runoff on June twenty-first. De la Espriella at forty-four, Cepeda at forty-one in the first round, with polling tightening to within two to three points. What to watch isn't just the outcome. Watch the margin, and watch whether it matches the polls. Emerging market desks have largely priced in a de la Espriella win — that's the consensus built into recent Colombian peso trading and energy equity notes.

A narrow de la Espriella victory is already in the price. A Cepeda upset — which the polling doesn't rule out — forces a repricing that's fastest in Colombian energy names and Latin American commodity funds. The second-order move: Chinese state-owned mining and infrastructure entities, which have methodically built supply relationships across South America, will recalibrate their Colombia strategy within forty-eight hours of any result.

In emerging markets, the surprise does more work than the consensus.

Wednesday, June twenty-fourth, after market close: Micron Technology reports its fiscal third-quarter earnings. Street consensus is at nineteen dollars and seventy-two cents per share on thirty-four point five billion in revenue. Micron guided fiscal Q3 to record revenue of thirty-three point five billion plus or minus three-quarters of a billion — a beat at current consensus is largely expected. Here's what actually matters: the fiscal fourth-quarter revenue guidance.

Micron has been the clearest real-time indicator of whether enterprise AI spending is translating into actual chip demand. The company ramped HBM3E — high bandwidth memory — aggressively this year, betting that AI model training and inference deployments would sustain that demand. If Q4 guidance holds or accelerates, the AI memory trade gets fresh validation. If the language around HBM orders becomes cautious, that's the tell.

What's already priced: a consensus beat. The market needs above-guidance Q4 guidance to move it meaningfully higher from here. A guidance miss — or even in-line guidance — triggers a sector-wide conversation. Memory is the semiconductor sector's most transparent demand indicator right now.

Micron doesn't just report earnings. It reports the state of the AI buildout in a language everyone can price.

Thursday, June twenty-fifth: the Conference Board releases May Consumer Confidence. Read past the headline number. In the last two Conference Board releases, the gap between the "Present Situation" component and the "Expectations" component has been widening — present conditions relatively resilient, expectations deteriorating. That divergence is the signal.

What's not fully priced: a crack in the "Present Situation" component. If current conditions start deteriorating alongside expectations, that's the early indicator for third-quarter consumer spending softening. The consequence: consumer staples outperform consumer discretionary. Small-cap domestic retailers — which carry the most direct exposure to domestic spending and the most leverage to tariff-driven price pressure — are where the repricing is fastest.

Two confidence numbers with the same name. One is what people think about now. One is what they think is coming. They've been moving in different directions.

Also Thursday: New Home Sales land. Follow it alongside the FHFA House Price Index, which lands Wednesday. What to watch specifically: whether new home sales growth is being driven by price cuts at the builder level or by rate-sensitive demand responding to the recent drop in mortgage rates toward six-and-a-half percent. Price-cut-driven volume is a supply story. Rate-driven volume is a demand story. Those are two different underlying realities dressed in similar words.

And Friday, June twenty-seventh — the main event. Three releases that collectively tell you the state of the US economy in a single morning: the final Q1 GDP revision, initial jobless claims, and May's PCE and core PCE inflation data.

Core PCE is the Fed's preferred inflation gauge. Wells Fargo economists expect it at zero point three percent month-over-month and three point four percent year-over-year. The market's current Fed funds futures pricing reflects roughly thirty percent probability of one rate cut by December 2026.

Here's the specific signal to watch: if core PCE comes in above zero point three percent, that hands Kevin Warsh the data confirmation he needs to begin setting up a rate hike. What's already priced doesn't accommodate a hike. Repricing for a hike from here hits long-duration Treasuries first, growth equities second.

If the Q1 GDP final revision goes downward from the advance estimate alongside a hot PCE print, you get the most dislocating combination possible: hot inflation in a slowing economy. That's the stagflation frame. Watch how the bond market reads the data in the first thirty minutes. The ten-year yield movement in that window is the real-time interpretation.

My personal watch-for this week is Wednesday night, when Micron reports. Not because of Micron specifically — but because the memory market is the cleanest real-time window into whether AI enterprise deployments are translating into actual, sustained chip demand. If the build is real, Micron says so. That answer comes Wednesday after the bell.

And that is how you wake up ready.

Knowledge Bomb

Here's the Knowledge Bomb: the early-career job market is not collapsing. It's mean-reverting.

That's less dramatic, which is why it's less popular. But it's what the data says.

The unemployment rate for twenty-to-twenty-four-year-olds is sitting around seven point two percent. The twenty-five-year average is about seven point three percent.

So the big scary chart isn't saying, "The labor market broke." It's saying, "Welcome back to normal."

And normal, for young workers, has always been a little ugly.

That's the first thing people forget. The entry-level labor market isn't the same as the adult labor market. It's the economy's shock absorber. When companies get nervous, they don't usually start by firing the experienced person who knows where all the passwords are. They start by slowing the hiring of people who still need training, onboarding, mentorship, and supervision.

So early-career unemployment moves first on the way down. Then it moves last on the way back up. That's the pattern.

In 2001, GDP growth slowed sharply after the tech bubble. Young-worker unemployment jumped almost immediately. In 2008, GDP barely had to turn negative before entry-level unemployment surged. And after the Great Recession, the economy technically recovered before young workers did. GDP came back. Corporate profits came back. The stock market came back. But entry-level hiring stayed miserable for years.

That's because companies don't re-open the front door just because the economy stops bleeding. First they work the existing staff harder. Then they restore margins. Then they approve budgets. Then they remember they used to have a campus recruiting team.

The pandemic labor market was the weird part. Not this. In 2021 and 2022, the economy went through a distorted reopening boom. Employers over-hired, wages jumped, remote work scrambled recruiting, and young workers briefly entered one of the strangest bargaining environments in modern history. Then rates rose. Capital got more expensive. Budgets tightened. And the twenty-to-twenty-four unemployment rate drifted back to its long-term average.

This isn't, by itself, proof that AI destroyed the first rung of the career ladder. It's not proof that Gen Z is uniquely doomed. It's proof that early-career labor is cyclical, volatile, and always more fragile than the headline unemployment rate makes it look.

The labor market didn't go from healthy to broken. It went from artificially tight to historically normal. And for young workers, historically normal has always meant: harder than advertised.

And that's the Knowledge Bomb.

Humor Me

It's two-seventeen p.m. in the Wealth and Means conference room. Three exhausted editors are hunched over laptops, surrounded by coffee cups, half-eaten takeout, and the dangerous belief that local news might save the segment.

The assignment is simple: find something light. Not another grim recap of world events. Just the good stuff from America's local newsrooms.

And then it happens.

St. George News drops a classic: "Highway patrol: Driver crashes into utility pole after reaching for candy that fell by her feet."

She reached for candy. The pole won.

No serious injuries, thankfully. But spiritually, the whole room stopped working. Because this is peak human performance. We have robots on Mars, AI writing code, markets trading overnight, and somewhere in Utah a grown adult looked down and said, "The Sour Patch Kids are on the floorboard. This cannot wait."

At that point, the meeting was basically over. Print it. Frame it. We're leading with the candy crash.

But local news doesn't quit after one masterpiece.

Boston.com had the MBTA general manager accused of pulling an employee's hair, which is not usually how you want transit leadership to make headlines. That's less "public infrastructure" and more "third-grade recess with pension obligations."

Santa Monica had a perennial political candidate arrested for grand theft. A man who ran for office more times than most people vote finally made headlines — just not in the brochure-friendly way.

And suddenly our standards were collapsing. We were supposed to be covering wealth, means, markets, incentives, and American enterprise. Instead, we were emotionally invested in whether the best sandwich shop in Arlington is the one with the giardiniera or the one with the Italian beef.

Then came Grundy Bank's Lip Sync Battle, which raised two thousand five hundred dollars for charity. That's the local news sweet spot: deeply wholesome, slightly chaotic, and exactly the kind of event where somebody's uncle performs "Livin' on a Prayer" like he's refinancing a boat.

Meanwhile, in Grand Junction, they were telling drivers to "zip up." Some mysteries are load-bearing.

And this is why local news matters. America's local editors are out there every day sifting through city commission swear-ins, hidden marijuana farms, pizzeria closings, transit drama, charity lip-sync battles, and public service announcements that sound like wardrobe advice. And somehow, they still find the perfect absurd gem for the showcase.

We walk in looking for inspiration and get reminded that half the country is one dropped Starburst away from becoming local lore.

So yes, we're leading with the candy lady. Obviously.

And if anyone asks what Wealth and Means stands for, tell them we invest in moments of pure, unfiltered human dumbassery. With love. Mostly with love.

And a gentle public service announcement: if the candy falls… just let it go.

The Greater Debate

Two lecterns. Heavy wood, low light — the kind of room where arguments are meant to last. On the left, a man built for crowds, for the raw machinery of American politics, for the belief that the federal government was the last institution big enough to take on suffering at scale. On the right, a man built for conviction, for the clean sentence that fits on a billboard and lives for forty years. Neither is here to concede. They've been having this argument for sixty years and neither has gotten tired of it.

Tonight's Greater Debate begins with one of the most dangerous phrases in American politics:

"We declared war."

Because once you declare war on something, people expect a victory parade.

In 1964, Lyndon Baines Johnson stood before the country and declared a War on Poverty. Not a program. Not a pilot. Not a grant initiative with a tasteful acronym. A war.

The question, sixty years later: Did we win?

To debate that, we bring together two presidents who never really stopped arguing with each other. On one side: Lyndon Johnson, architect of the Great Society, believer in federal power, and the man who insisted that poverty was not just personal misfortune, but a national failure. On the other: Ronald Reagan, apostle of limited government, critic of the welfare state, and the man who delivered one of the most brutal postmortems in modern politics:

"In the sixties we waged a war on poverty, and poverty won."

That's not a policy critique. That's a tombstone inscription.

Johnson opens with the moral case.

America is not great because it leaves people alone. America is great because it refuses to leave people behind.

To LBJ, the War on Poverty was never supposed to be a hammock. It was supposed to be a ladder. Not a handout. A hand up.

He points to Job Corps, Head Start, work-study, community action, Medicaid, Medicare, education funding — all as pieces of the same argument. You cannot tell a child to compete in the American economy if that child starts life hungry, sick, segregated, undereducated, and medically invisible.

In 1959, the official poverty rate was twenty-two point four percent. By 1969, it had fallen to twelve point one percent. By 1973, it hit eleven point one.

That's not failure. That's the biggest poverty reduction in modern American history.

Johnson argues the War on Poverty worked when America actually fought it. Then came Vietnam. Then budget pressure. Then backlash. In LBJ's telling, America didn't lose the War on Poverty because the strategy failed. America lost interest.

Now Reagan leans forward.

Lyndon. You're counting the wrong thing.

You're counting spending. You're counting programs. You're counting federal intention. But the true test of compassion isn't how many people receive government help. It's how many people no longer need it.

Reagan argues that the War on Poverty misdiagnosed poverty as a shortage of government activity when, in his view, the deeper engines of mobility were work, marriage, family stability, private enterprise, and local responsibility.

If the War on Poverty worked, why did the official poverty rate stop falling?

After the early 1970s, the official poverty rate mostly moved in a narrow, cyclical band — rising in recessions, falling in expansions, but never permanently collapsing toward zero. Meanwhile, spending exploded. A modest mid-1960s safety net became a trillion-dollar-plus modern anti-poverty apparatus.

Reagan's cross-examination is simple: if we spent more and more, why didn't the official poverty rate fall more and more?

He says the answer is dependency. Welfare programs, however compassionate in intent, can distort incentives in practice. Phaseouts can punish additional work. Benefit rules can penalize marriage. A bureaucracy can become more interested in managing poverty than eliminating it.

Johnson hears this and snaps back.

You're confusing the ambulance with the accident.

If someone is poor and receives food assistance, the program didn't create hunger. If a mother receives support, the benefit didn't create the low wage, the absent employer healthcare plan, or the broken school system. Reagan's argument turns the safety net into the suspect simply because it's found at the scene of suffering.

And here the debate becomes less about poverty and more about human nature.

Johnson believes government can expand opportunity. Reagan believes government often replaces responsibility. Johnson sees poverty as a structural barrier. Reagan sees poverty policy as an incentive system.

Johnson asks: What does a decent nation owe its citizens?

Reagan asks: What happens when the state provides what families, markets, churches, neighborhoods, and individuals once had to provide?

Now the modern evidence complicates both men.

Because the official poverty rate tells one story. The Supplemental Poverty Measure tells another. The old official measure is basically a cash-income test built from a 1960s food-budget formula. It doesn't fully count many of the modern benefits government now provides. If SNAP, housing assistance, Medicaid, refundable tax credits, and other supports reduce hardship but aren't counted in the official number, then the scoreboard is broken.

That's powerful evidence for Johnson. Government transfers can reduce measured hardship. Quickly.

But it's also evidence for Reagan. If poverty reduction depends on constant transfer expansion, is that victory — or maintenance?

The honest synthesis: Johnson was asking what a rich country owes to the people its systems failed. His answer was ambitious, federal, and built on the belief that structural disadvantage required structural remedy.

Reagan was asking what happens to agency, habit, and self-sufficiency when government removes the consequences of inaction. His answer was smaller, local, and built on the belief that the welfare state, however well-intentioned, can slowly replace the very capacities it was meant to restore.

They're not the same question. And that's exactly why this debate hasn't ended.

The War on Poverty was not won, because poverty still exists. It was not lost, because millions of people are materially better off because the safety net exists. But it remains unfinished because America never fully resolved the contradiction at the center of the project: we want government strong enough to prevent destitution — but not so strong that it replaces the habits, institutions, and incentives that help people become self-sufficient.

LBJ wanted a Great Society. Reagan wanted a free society. The unresolved work is building one that can be both.

And that… is The Greater Debate.

Let's Invent Again

In 1949, if you were building a computer, you had a memory problem. Not a metaphorical one. A literal one.

The machines of that era stored data in cathode ray tubes — temperamental glass cylinders that required constant power to maintain their contents. If the power flickered, the data was gone. If the tube degraded, the data was corrupted. You could build a machine that was brilliant at calculation, but you couldn't reliably make it remember what it had just calculated. The early computers were, in a sense, very fast amnesics.

The problem was hiding in physics. And a twenty-nine-year-old man who'd arrived in the United States from Shanghai four years earlier was about to find it.

His name was An Wang. He'd earned his bachelor's degree in science from Chiao Tung University in Shanghai in 1940, come to America in 1945 to study further, and picked up his PhD in applied physics from Harvard in 1948. He then stayed at the Harvard Computation Laboratory — the place where the early machines were being built and puzzled over — and started working on the memory problem.

What he discovered in 1949 was the principle of magnetic core memory. The idea was elegant once you saw it. Take a tiny ring — a "core" — made of ferromagnetic material, about the size of a pinhead. Thread a wire through it. Send a pulse of electricity through the wire and the core magnetizes in one direction — that's a one. Send a pulse the other way and it magnetizes in reverse — that's a zero. Read the state with a second wire. And here's the crucial part: the core holds its magnetic state without any power at all. Turn the machine off. Come back tomorrow. The data is still there, encoded in the physical orientation of thousands of tiny rings, waiting quietly in the dark.

An Wang received the patent in 1950. In 1956, he sold it to IBM for four hundred thousand dollars.

That transaction would be debated for decades as one of the most undervalued patent sales in American history. IBM built magnetic core memory into the backbone of the computing industry. For roughly twenty years — from the mid-1950s through the mid-1970s — it was the dominant form of computer memory in the world. Military systems. Banking computers. Airline reservation systems. The first real-time computing networks. All running on the principle An Wang had proved in a Cambridge laboratory.

And then there's this.

In 1969, the Apollo Guidance Computer — the machine that navigated astronauts to the moon and back — ran on magnetic core memory. The system that calculated orbital trajectories, managed fuel burn, and guided the lunar module to the surface of the moon stored its data in the magnetic orientation of tiny rings threaded with copper wire, each one embodying the principle An Wang had discovered twenty years earlier.

The moon landing ran on An Wang's invention.

He used the four hundred thousand dollars from the IBM sale to fund Wang Laboratories. The company he built would go on to pioneer electronic calculators and then dominate the office word processing market in the 1970s and 80s. The Wang word processor was so omnipresent in American offices that the phrase "doing the Wang" entered business vernacular the same way "Googling something" did a generation later. He held over thirty-five patents by the time he died in 1990. He was inducted into the National Inventors Hall of Fame in 1988.

But the deepest thing An Wang left behind isn't measured in patents or market capitalizations. It's measured in the idea that information could be encoded in physical states — invisible, persistent, reliably readable. That a machine didn't just have to calculate. It could remember.

Every piece of technology that stores something — every phone, every server, every cloud drive — owes something to the principle he found in a tiny magnetized ring in 1949.

He arrived in America with a physics degree and an idea about how matter could hold information. He changed what computers could be. And somewhere between a Harvard lab and the surface of the moon, he proved that the most powerful signals are often the ones you can't see at all — until you know exactly how to read them.

Chapters