On US Politics: Nearly Everything You Need to Know

The claim. Markets create wealth, wealth spreads, and everybody ends up better off than they would have been otherwise. Self-interest at the top produces good at the bottom. That is the deal, and it has worked better than anything else anyone has tried.

My answer. That is the sales pitch. It was never the goal.

The actual dream of unfettered capital is unlimited individual accumulation with no obligation to the collective. No living wage, no healthcare for workers, no subsidised childcare, no food security, no share of the surplus. The capitalist resists every policy that would make his employees’ lives easier, including the ones that would make them more productive, because the point was never efficiency. The point is to keep everything. Prices up and wages down, not as an unfortunate byproduct but as the explicit aim.

The argument that self-interest produces collective good depends on two things staying true: competition holding prices down, and labour scarce enough to be able to demand fair pay. Capital consolidates and competition dies. Labour is kept desperate through policy. Both legs were kicked out by the same people who make the argument.

Collective growth is not a promise unfettered capitalism failed to keep. It was never made by the people actually holding the capital. It was made about them, by economists and politicians paid to make extraction sound like a public service. The dream works as ideology: a story the capitalist tells himself so he can feel virtuous while he extracts.

And the other half of it. Nobody earns a billion dollars. You can work hard. You can be brilliant. You can take enormous risks. No human being, through their own labour, produces a billion dollars of value. That figure is only reachable by taking the surplus of thousands of other people’s work and calling it your own. The point is not to abolish ambition. It is to abolish the mechanism by which one person’s ambition requires another person’s poverty.

The claim. Employers worked out on their own that rested workers are better workers. Conditions improved because prosperity improved. Nobody had to be forced, and the labour movement mostly took credit for what growth was doing anyway.

The record. Every one of those was taken, not given, and the bill was paid in bodies.

Robert Owen coined "Eight hours labour, eight hours recreation, eight hours rest" in 1817. Sixty-nine years later Americans were still dying for it. The Haymarket affair of 1886 killed at least eleven people, seven of them police officers, and four anarchists were hanged for it afterwards. The Homestead strike of 1892 put three hundred armed Pinkerton agents against workers at Carnegie Steel and left seven workers and three Pinkertons dead. Between 1866 and 1892 the Pinkertons worked seventy labour disputes against more than 125,000 strikers. Federal troops and a federal injunction broke the Pullman strike in 1894. At Ludlow in 1914 the Colorado National Guard fired on a tent colony of striking miners and killed at least nineteen people, including eleven children and two women who suffocated in a pit dug beneath the tents to hide them from the shooting.

Henry Ford put his factory floor on a five-day, forty-hour week in 1926, and he was explicit that it was not kindness. Tired workers make mistakes and buy fewer cars. The Fair Labor Standards Act of 1938 set a forty-four-hour week, dropping to forty by 1940, along with the first federal minimum wage that survived the courts, at twenty-five cents an hour. The same act carried the first durable federal restriction on child labour. Before it, the 1870 census counted more than 750,000 workers under the age of fifteen, and by 1900 more than one American child in five was at work.

The weekend. Overtime. Workplace safety. Every protection you take for granted was paid for by people your history books left out. These were not inevitable developments. They were concessions extracted from capital under threat of revolution.

And capital’s willingness to use force against a democracy is not ancient history either. In 1933 and 1934 a group of wealthy businessmen approached the retired Marine general Smedley Butler about overthrowing Franklin Roosevelt and installing a government modelled on Mussolini’s Italy. Butler, who had spent his career fighting wars for corporate interests and later wrote War Is a Racket, reported it to Congress. The McCormack-Dickstein Committee found that the plot had been discussed and planned and might have been put into execution had the financial backers thought it expedient. It named no conspirators publicly. Nobody was prosecuted. The links to DuPont, General Motors and J. P. Morgan interests were alleged by Butler and never established by the committee. That is documented congressional testimony, not a conspiracy theory, and it has been almost entirely forgotten. When the New Deal threatened accumulated wealth, some of the wealthiest men in America listened to a pitch for a coup.

The claim. Growth comes from letting people keep their own money. The fifties were the golden age, and the golden age was built by getting government out of the way.

The record. The tax code tells you who a society actually values, and the golden age had the highest rates in American history.

From 1942 to 1963 the top statutory marginal income tax rate never fell below 82.13 percent. At its peak, in 1944 and 1945, it was 94 percent on income above $200,000 — about $3.8 million in today’s money. It held at 91 percent from 1954 to 1963, seven of those years under Eisenhower, a Republican president modern conservatives claim as a saint. Corporate tax sat at 52 percent from 1952. The estate tax topped out at 77 percent on the largest estates from 1941 to 1976.

And the economy boomed. Median family income more than doubled between 1947 and 1973. The poverty rate fell from about a third of the country in 1948 to 11.1 percent by 1973. The 41,000 miles of the Interstate Highway System were authorised under Eisenhower in 1956. The G.I. Bill sent 2.2 million veterans to college and guaranteed the mortgages that built the American suburb. The United States built its physical and intellectual infrastructure out of revenue taken from accumulated wealth, and it worked. Not because the wealthy were generous. Because they had no choice.

Even the first cut was modest by anything we would recognise now. Kennedy proposed bringing the top rate from 91 to 65 percent in January 1963, arguing it had grown high enough to distort behaviour. After his assassination the Revenue Act of 1964 set it at 77 percent for that year and 70 percent from 1965. At 70 percent the country still built Medicare and Medicaid in 1965, funded the Great Society, and kept real wages rising. High marginal rates on the wealthy did not destroy the economy. They paid for it.

Look at what else was true in those years, because the tax rate was not doing this on its own. Union membership was climbing toward the all-time high of 34.8 percent it reached in 1954, on the Bureau of Labor Statistics series. Wages compressed hard: the gap between the top tenth and the bottom tenth of earners fell by about a quarter across the 1940s and stayed compressed until the 1970s. Productivity and the pay of a typical worker rose together from 1948 to 1973 and then stopped. And the public spent enormously on people: 7.8 million veterans went through the G.I. Bill, the National Defense Education Act put over a billion dollars into loans and fellowships in 1958, and California’s Master Plan of 1960 made its universities tuition-free for residents until Proposition 13 ended it. That is a country running a bargain, not a market left alone. I will state my position on it as mine: the America people are nostalgic for was running its own version of what we now call the Nordic model.

Eisenhower’s own warning, in his farewell address of January 1961, was not about taxes at all. It was about the military-industrial complex, the corporate-military entanglement that has since eaten most of the federal budget. He had watched it build for eight years. He named it on his way out the door. Nobody listened.

The claim. Money left at the top gets invested, investment makes jobs, jobs make wages. A rising tide lifts all boats. Supply-side.

The record. The man running the policy said in print, while it was being passed, that the economics was cover.

Arthur Laffer sketched his curve over dinner for Dick Cheney in late 1974. The famous napkin in the Smithsonian is dated September and inscribed to Donald Rumsfeld, and both the date and Rumsfeld’s presence are disputed; Laffer’s own recollection is that the original went in the bin. The curve is not wrong in the abstract. There is some rate at which taxation becomes self-defeating. The trick was applying that logic at rates nowhere near it, as mathematical cover for what was a transfer. The curve gave the con a graph, and the graph gave the con respectability.

Reagan’s Economic Recovery Tax Act of 1981 took the top marginal rate from 70 to 50 percent. The Tax Reform Act of 1986 took it to 28, the lowest since 1931. And David Stockman, Reagan’s own budget director, told the journalist William Greider in The Atlantic in 1981, on the record, that "Kemp-Roth was always a Trojan horse to bring down the top rate." In plain English: the economic argument was the pitch, and the rate cut was always the point. Stockman was not fired. He ran the Office of Management and Budget until 1985 and the policy continued. George H. W. Bush had called it "voodoo economic policy" while running against Reagan in April 1980, then became his vice president and helped implement it anyway.

What followed was not investment. It was extraction. The national debt went from $998 billion at the close of the 1981 fiscal year to $2.9 trillion at the close of 1989, because the promised revenue growth never arrived. The gap was papered over with deficit spending and the bill forwarded to future workers. The rich kept their cuts. Everyone else got the debt.

The estate tax tells the same story. It stood at 70 percent above $5 million in 1977. Reagan cut it, Bush cut it further, and the Tax Cuts and Jobs Act of 2017 doubled the exemption to roughly $11.2 million per person and $22.4 million for a married couple before any tax applies at all. Once a brake on dynastic wealth, it now works more as a public relations concern for the very rich than as a fiscal mechanism. The aristocracy the founders claimed to oppose has been quietly rebuilt, inheritance shielded behind trusts and exemptions, while people who work pay full rate on every dollar they earn.

And it was tested at home, in my own state. In June 2012 Governor Sam Brownback went on television and said of his Kansas tax cuts, "we’ll have a real live experiment." Revenue collapsed. Education funding was gutted, infrastructure maintenance deferred, deficits ran year after year, and in June 2017 a Republican-controlled legislature repealed the cuts over his veto. Five years, and it failed by every measure anyone had agreed to measure. The supply-siders called it a flawed implementation. They always do. The theory is never wrong. The execution is always the problem.

Four decades, several countries, many implementations, and not once has it produced the promised broad-based growth. What it reliably produces is higher after-tax income at the top, flat wages below, underfunded public services, and political power transferred to the people who can now afford to buy it. The con required Americans to believe that men freed from any obligation to contribute would choose to share anyway. They were never going to. The history is right there, from 1945 to 1980, when they were made to.

And it has been tested by people whose job is testing it. Zidar, in the Journal of Political Economy, finds the employment effect of tax cuts comes almost entirely from cuts for lower earners, and that cutting taxes on the top ten percent moves employment growth very little. Hope and Limberg went through eighteen rich countries across fifty years and thirty separate major tax cuts for the rich, and found the effects on growth and on unemployment statistically indistinguishable from zero — with one thing moving reliably: the top one percent’s share of pre-tax income, up more than seven tenths of a percentage point within five years. There are serious economists on the other side, Mertens and Montiel Olea the best of them, and it is not a closed question. It is nothing like as closed as it is sold.

Start with what happened, because it is recent and it is measured.

In February 2026 the Supreme Court held 6 to 3, in Learning Resources, Inc. v. Trump, that the emergency-powers statute the tariffs were built on does not authorise tariffs at all. The Chief Justice wrote for the Court that the law contains no reference to tariffs or duties, and that laying a duty is a power the Constitution hands to Congress. The government has been paying the money back ever since: 104.29 billion dollars authorised in refunds, 71.06 billion paid as of the end of June 2026, with roughly a hundred billion still owed.

What remains in force is narrower and rests on older statutes: steel, aluminium and copper at 15 to 50 percent, autos at 25, pharmaceuticals at 100, and a set of country tariffs running 10 to 12.5 percent. The average rate actually collected in June 2026 was 6.8 percent of import value, against 2.3 percent in 2024.

Who paid it. Four separate studies, the New York Fed, Brookings, Harvard, and the IMF, converge on the same answer: about 90 percent of the tariff burden fell on American importers, not on foreign exporters. The Peterson Institute's August 2026 review of the whole literature puts it plainly: US consumers and firms together bore about 90 percent. No study found otherwise.

What it bought. American manufacturing employed 19,553,000 people at its peak in June 1979. It was 17,181,000 in December 2000 and 11,453,000 by February 2010, 5.7 million jobs gone in a single decade, most of them before anybody in Washington was arguing about tariffs. In August 2025 the figure was 12,672,000. In August 2026 it was 12,707,000. Thirty-five thousand jobs over a year, after a decline through 2025 and a recovery through 2026.

The composition is the interesting part. Primary metals and fabricated metals (the protected industries) added jobs. Machinery, computers and transportation equipment, which buy metal as an input, lost them. Protection worked for the protected and was paid for by everybody downstream of them.

And the trade deficit, the thing the tariffs were meant to close, went up in the first tariff year: from 885.5 billion dollars in 2024 to 932.0 billion in 2025.

Here is something worth knowing about the tariff code that already exists. It is written backwards. Cheap stainless steel spoons are taxed more heavily than sterling silver ones. Infant formula is taxed more heavily than champagne. Across apparel, footwear and household goods, the low-cost version routinely carries the higher rate. The poorest tenth of households pay around 95 dollars a year in tariffs and the richest tenth around 500 dollars, which sounds proportionate until you notice the richest tenth earn many times more than five times as much. Measured against income, the tariff is one of the most regressive taxes in the American system, and it has been for a long time without anybody campaigning about it.

My position, stated as mine: a trade deficit only costs you while you are not making the things. It is a description of dependence, not a debt. Fix the dependence and the number stops mattering. Once a country makes most of what it actually needs, a tariff stops being a tax on the working family's groceries and becomes something else entirely, a wealth tax on exotic imports, paid by the people buying the things nobody needs. That is a good idea in a country that manufactures. It is a bad idea in a country that does not, which is the country we are, which is why it is landing the way it is landing.

What the record does show about building an industry deliberately: the best-identified case in the economics literature is South Korea's heavy and chemical industry drive from 1973 to 1979, studied by Nathan Lane in the Quarterly Journal of Economics in 2025. It worked, the gains spread to the industries downstream of the targeted ones, and they outlasted the policy. It was narrow, it was conditional, the firms had to hit export targets to keep the support, and it had an end date.

The claim. Force wages up, force the surplus down, or break up the winners, and you kill the growth that pays for all of it. The money has to be left at the top and allowed to find its own way down.

The record. Some countries stopped waiting for it to find its own way and wrote the route into law. Their capitalism came out more competitive than ours, not less.

Start with the finding that surprised me most, because it is not what either side of this argument usually says. Mogstad, Salvanes and Torsvik went through Nordic and American income data for the National Bureau of Economic Research last year. Market income inequality — before a single tax is collected or a single benefit paid — runs at 0.39 on the Gini scale in Norway and 0.40 in Denmark. The United States is 0.47 before taxes and transfers, and still 0.39 after them. Norway’s raw market outcome is already where America lands after everything its government does. In their words, a more equal predistribution of earnings, rather than redistribution, is the main reason for lower inequality in the Nordic countries. They are not taxing their way there from an American starting point. The money arrives distributed.

Here is the machinery that distributes it.

The wage floor is bargained for a whole industry at once. Denmark, Sweden, Norway, Finland and Iceland have no statutory minimum wage at all. The floor comes out of sectoral agreements, and the agreement covers the sector whether you joined anything or not: 82 percent of Danish workers are covered by one, 88 percent of Swedes, 89 percent of Finns, 72 percent of Norwegians. In the United States the figure is 11 percent. A Danish hotel and restaurant worker is on roughly 141 to 176 kroner an hour under the current agreement — call it twenty dollars and up — with pension, six weeks of holiday and shift supplements written into the same document. No legislature set that wage. A union set it for the entire industry in one negotiation.

Workers sit on the board. Sweden requires worker representation on company boards at twenty-five employees, Norway at thirty, Denmark at thirty-five. Germany gives workers a third of the supervisory board at five hundred employees and half of it at two thousand. The United States requires it at no number of employees whatsoever.

The rent gets caught before it becomes private wealth. Norway taxes petroleum at a 78 percent marginal rate and puts the proceeds into a fund now worth about 22.7 trillion kroner — roughly 3.8 million kroner for every Norwegian alive — under a rule that the state may spend the expected real return and never touch the capital. More than half of that fund is now investment return rather than oil money. The oil was under the seabed of a capitalist country with a stock exchange. What differed was whose name went on the deed before the drilling started.

The churn gets paid for instead of denied. Denmark’s firing rules are about as liberal as Britain’s and Danish workers change jobs about as often, because the security is attached to the person rather than to the job. Denmark spends about 1.2 percent of its economy moving people between jobs. The United States spends about a tenth of one percent.

Now the competition half, which is the part nobody expects. American antitrust changed its mind at the end of the 1970s: Bork’s The Antitrust Paradox in 1978, the Supreme Court adopting the consumer welfare standard in 1979, and enforcement following behind. Kwoka went through the Federal Trade Commission’s own merger investigation files. Mergers that would leave five significant competitors standing were challenged 40 percent of the time in the late 1990s, and zero percent of the time from 2008 onward. Not less often. Zero.

Europe went the other way, and the reason is worth knowing. Gutiérrez and Philippon’s account is that when the member states built a shared competition regulator, each of them feared the others capturing it more than it wanted to capture it itself — so between them they made it more independent than any of them had ever made its own. The result is a regulator enforcing free markets harder than any single member government would choose. Their measurements: American concentration rose across most sectors after 2000 while European concentration stayed flat or fell; American profit margins rose while European ones did not; American home broadband ran about 68 dollars a month against 30 to 40 euros in western Europe. Philippon’s own framing is not that Europe is less friendly to markets. It is that Europe kept enforcing the free market after America stopped.

And they do still enforce it. Norway’s competition authority fined the country’s three grocery chains 4.9 billion kroner in 2024 for swapping price information through employees sent into rivals’ stores with hand scanners, and the appeals tribunal upheld it in full the following year. The European Commission has fined Google 2.42 billion euros over shopping, 4.34 billion over Android — upheld this July — and 2.95 billion over advertising technology last September. Under the Digital Markets Act it fined Apple 500 million euros and Meta 200 million in April 2025, with the power to go to ten percent of worldwide turnover and, in the last resort, to order a break-up.

What is disputed, and it is a fair amount. Whether American concentration really rose depends on how you draw the market: national concentration went up on most measures while local concentration went down, because the big chains moved into towns that previously had one of everything. The most-cited evidence for rising American markups was substantially undermined by a replication last year which found the original had dropped about a quarter of its sample without saying so. The Information Technology and Innovation Foundation measures concentration at a finer industry level and finds barely any American increase at all. And Kleven’s argument for why Scandinavians can tax so much is that it rests on machinery — near-total third-party income reporting, a Danish evasion rate around 2.2 percent — that a country cannot vote itself overnight.

None of that touches the first number. Before tax, before transfers, before any of the arguing about who deserves what, their market hands people a more equal result than ours does after we have finished trying to fix ours.

The claim. Competition disciplines companies. Regulation is what breaks markets. The answer to a captured system is to shrink the thing doing the capturing.

The record. The market was captured on purpose, to a written plan, and shrinking government is the plan working as intended.

The intellectual groundwork was laid in 1947, when Friedrich Hayek gathered thirty-six economists, philosophers and journalists at Mont Pelerin in Switzerland. Milton Friedman and Ludwig von Mises were among them. For thirty years their ideas sat on the fringe, funded and developed and waiting. Then the stagflation of the seventies discredited the Keynesian consensus in the eyes of policymakers, and Thatcher and Reagan implemented the Mont Pelerin programme wholesale: deregulation, privatisation, union-busting, tax cuts at the top. The ideas were not new in 1980. They had been waiting for a crisis that would make them saleable.

On 23 August 1971, two months before Nixon nominated him to the Supreme Court, the corporate lawyer Lewis Powell sent a confidential memorandum to the U.S. Chamber of Commerce titled "Attack on American Free Enterprise System." It set out a coordinated strategy for business to capture the academy, the press, the courts and the political process, and told corporations to stop behaving as individual competitors and start behaving as a single political force. Fund think tanks. Place business-friendly academics. Monitor textbooks. Build a legal apparatus to challenge regulation. The Heritage Foundation followed in 1973, the Cato Institute in 1977, the Federalist Society in 1982. The counterrevolution was not a drift. It was a blueprint.

Then look at what competition actually produced. The six largest American bank holding companies held assets equal to about 15 percent of GDP at the end of 1994 and about 62 percent by the end of 2009. The fifty largest retail firms went from 25.7 percent of retail sales in 1997 to 33.3 percent by 2007. Walmart alone takes about seven percent of all American retail spending and about a fifth of the grocery market. A handful of conglomerates make most of the packaged goods on the shelf. Different labels. Same owners.

The economist Glen Weyl put it about as plainly as it can be put: "It says the very structure of capitalism, the system of private property itself, is inherently monopolistic." Competition does not prevent consolidation. It selects for it. The winner of the competition uses the winnings to buy the competition. Then the regulators. Then the legislators.

That last step has a name. Regulatory capture. The agencies built to protect you from corporations are staffed by people who came out of those corporations, who write rules their former employers can live with, and who return to the private sector for the reward. The system is not broken. It is working exactly as designed. Just not for you.

Meanwhile the union membership that used to be the counterweight fell from a peak of 34.8 percent in 1954 to roughly one worker in ten. Since 1979 net productivity has grown 93 percent while the pay of a typical worker has grown 34 percent — productivity rising nearly three times as fast as compensation, on the Economic Policy Institute’s numbers to the first quarter of this year. Before 1979 the two lines moved together. When workers produced more, they earned more. The decoupling is precise and documented. The surplus did not disappear. It was redistributed upward.

That decoupling tracks the dismantling of the tax structure almost exactly, and the mechanism is not mysterious. When an executive’s take-home is capped by a 91 percent top rate, there is limited reason to maximise your own compensation at your workers’ expense, because the marginal dollar above the threshold mostly goes to the government anyway. Drop that rate to 28 and every dollar moved from wages to executive pay is a dollar the executive keeps. The incentive changed. The behaviour followed. That is arithmetic.

The claim. Whatever its faults, the market leaves you free. A government bureaucrat is a master you did not choose. An employer is a choice you can walk away from.

My answer. What control? The control to choose which employer underpays you? Which insurance company denies your claim? The freedom to go bankrupt over a hospital bill, or to work until you die because retirement is not affordable? That is not freedom. That is choosing which master holds the leash.

Look at where the money goes. Before 1982 share buybacks were rare, because a company repurchasing its own stock ran a serious risk of a market-manipulation charge. Then the SEC under Reagan adopted Rule 10b-18 and handed them a safe harbour. In 2018 alone the companies of the S&P 500 spent $806 billion buying back their own shares — money that could have gone to wages, research or expansion, spent instead inflating a share price that executive compensation is tied to. That is not investment. That is extraction.

Retirement went the same way. In 1980, 38 percent of private-sector workers took part in a defined-benefit pension, an income guaranteed for life. By 2025 only 14 percent even had access to one. The 401(k) that replaced it entered the law in the Revenue Act of 1978 and the first plans opened in 1982. It shifts every risk onto you, charges management fees that compound across forty years, and leaves you with whatever the market happens to be worth in the year you stop. Risk transferred. Security dissolved. Policy, not accident.

And you are likelier to be robbed by your employer than by a stranger. The Economic Policy Institute puts minimum-wage violations alone at about $15 billion a year. On its own earlier count, the wages actually recovered for American workers in 2012 came to $933 million, against $340.9 million taken in every robbery in the country that year. The difference is what happens next. Wage theft is rarely prosecuted, and when it is, the penalty is a civil fine. The worker who takes from the register goes to prison. The employer who takes from the worker pays and keeps trading.

The federal minimum wage has been $7.25 since 2009. That is seventeen years without an increase, the longest stretch since the minimum wage was established in 1938, and its lowest real value in seven decades.

You say you will not hand your rights to a government that does not care about you. I agree. But you have already handed them to corporations that do not care about you either, and at least a government can theoretically be voted out.

The claim. Every attempt ends in queues, shortages and secret police. The examples are all around you and they all point the same way.

The record. You are being shown the cherry-picked failures and kept away from the scoreboard.

For every Venezuela there is a Norway. For every Cuba there is a Denmark. The difference is rarely the presence or absence of socialist policy. It is whether the country was sanctioned, couped or invaded by the United States for nationalising its own resources.

Now take the rankings, including the ones kept by people who would much rather they said something else. Finland has come first in the World Happiness Report for nine years running; the United States is 23rd. Freedom House scores Finland 100 out of 100, Norway and Sweden 99, Denmark 97, and the United States 81 — which this year put America outside the global top twenty for the first time. The Economist Intelligence Unit has classified the United States a flawed democracy rather than a full one since 2016, and its 2025 index, published this April, ranks the country 34th, down from 28th. And the Heritage Foundation, about as free-market an outfit as this country has, puts the United States 22nd on its own 2026 Index of Economic Freedom, behind Denmark at 7th, Norway at 8th, Sweden at 11th and Finland at 13th.

None of those countries are socialist in the Soviet sense. They have market economies. They also have universal healthcare, free education, strong unions, generous parental leave and real safety nets, funded by progressive taxation. They prove you can have markets without having poverty, and entrepreneurship without desperation. By capitalism’s own metrics, kept by capitalism’s own institutions, the social democracies are winning.

And these are not only satisfaction surveys. They are body counts. American life expectancy reached an all-time high of 79.0 years in 2024. A Norwegian can expect 83.2. American infant mortality was 5.53 per thousand live births in 2024; Finland’s runs around 2.1. American maternal mortality was 17.9 per hundred thousand births in 2024, improved from 22.3 in 2022 and still the worst in the developed world, roughly triple the rate in Sweden, Japan, the Netherlands, Germany, the United Kingdom or France. American women are more likely to die in childbirth than women in any other wealthy country. That is not an accident. That is policy.

The claim. Nordic policy runs on a level of mutual trust that only a small and similar population can produce. America is too large and too varied for any of it to transfer.

My answer. Japan has 125 million people and universal healthcare. Germany has 83 million and free university education. The homogeneity argument is a dogwhistle, and what it means is that we cannot have nice things because we would have to share them.

The United States has already run the experiment. The most equitable this economy has ever been was the stretch with the highest union membership and the highest marginal tax rates. That is the same period conservatives now call the golden age they want to return to, without ever acknowledging what made it golden.

And the argument is usually run backwards. Small is not the advantage. My position, stated as mine: the Nordic model with the natural resources of the United States or China behind it would be the dominant economy on earth, and what those countries are working with is a fraction of what we are sitting on. Norway built a fund worth 3.8 million kroner a head out of one oil province in the North Sea.

The claim. Argue about fairness all you like: America is the most competitive economy in the world, and the countries doing it some other way are not really in the race.

My answer. China is beating us at our own game, and the two places you can see it clearest are the trade balance and the concrete.

Trade first. China’s goods trade surplus was 992 billion dollars in 2024 and 1.19 trillion in 2025, the largest any country has ever run. Through August of this year it is at about 806 billion and on pace to beat that again. The American goods deficit in 2025 was 1.24 trillion. Tariffs did not stop it. In 2025 China’s exports to the United States fell about a fifth, and roughly 150 billion dollars of goods simply went somewhere else — 65 to 70 billion of it into southeast Asia, 25 billion into sub-Saharan Africa — leaving the global surplus intact. Then the Supreme Court struck the tariffs down in February, and in August alone China’s exports to the United States were up 34 percent on the year before.

China makes about 28 percent of the world’s manufactured value added. The American share is somewhere between 15 and 17 percent, down from about a quarter in 2000. Three quarters of the world’s electric cars are built there. More than 80 percent of the battery cells, about 85 percent of the cathode material, more than 90 percent of the anode. 98 percent of the world’s solar wafers. More than 60 percent of all new shipbuilding orders placed in 2025 — and the American share of the commercial shipbuilding market is one tenth of one percent.

Now the concrete. China has more than 50,000 kilometres of high-speed rail, more than the rest of the world put together, twelve thousand of it laid in the last five years, reaching 97 percent of its cities over half a million people. The United States has none: the Acela runs on shared legacy track and averages about seventy miles an hour end to end. Six of the world’s ten busiest container ports are Chinese; the busiest American one, Los Angeles, sits sixteenth and moves about a fifth of what Shanghai moves. China added 430 gigawatts of wind and solar last year. The United States added 53 gigawatts of everything combined. On the McKinsey Global Institute’s numbers, China’s net productive capital investment in 2024 ran about eight times America’s. The American Society of Civil Engineers graded American infrastructure a C in 2025 — its best grade ever awarded — and put the ten-year investment gap at 3.7 trillion dollars.

What is disputed, and it is substantial. Chinese household consumption is 40 percent of that economy against America’s 68, which is another way of saying the surplus is financed by what Chinese households are not allowed to spend. Property investment fell 17 percent last year, total fixed asset investment fell 3.8 percent, industrial capacity use is at a six-year low, prices have been flat or falling for ten straight quarters, and Rhodium puts real growth at 2.5 to 3 percent against the official 5. American output per head is six and a half times China’s. Lu Dadao, a Chinese Academy of Sciences academician rather than a foreign critic, argues that 80 to 85 percent of that rail network loses money, on top of about 855 billion dollars of railway debt. And the strongest form of the objection is this: a state spending four times America’s share of its economy on industrial subsidy, holding equity in something like 867,000 firms and setting output quotas has not out-competed American capitalism at all. It has shown that a different system can out-produce one. Take that seriously. It is the best argument against what I just said.

The claim. Conservative states carry themselves and subsidise the rest. The welfare and the dysfunction are somewhere else.

The record. It runs the other way, and it has for a long time.

The Rockefeller Institute of Government works out, every year, what each state pays the federal government and what it gets back. The national average in federal fiscal year 2023 was 1.32 dollars received for every dollar paid.

The states that get the most back: New Mexico 2.91. Mississippi 2.57. West Virginia 2.57. Kentucky 2.28. Alaska 2.26. Virginia 2.22. Alabama 2.18. Louisiana 2.00. Maryland 1.92. Oklahoma 1.91.

The states that get the least: New Jersey 0.88. Massachusetts 0.95. Washington 1.00. New Hampshire 1.00. California 1.02. New York 1.04. Minnesota 1.06. Illinois 1.06.

Say the honest version of this, because the honest version is stronger than the slogan. The single most subsidised state in the country is New Mexico, and New Mexico votes Democratic. Virginia and Maryland are high on the list because the federal payroll is next door to them, not because they are poor. Texas at 1.21 and Florida at 1.18, the two biggest states that make this argument loudest, are both below the national average. Roughly a third of the states sit on the opposite side of the average from where their politics would put them.

What holds without exception is the bottom of the list. The states that put in more than they take out are, almost to a state, the ones that get called takers.

And what is actually moving is not welfare. About 62 percent of federal expenditure in those figures is direct payments to individuals, overwhelmingly Social Security and Medicare. Another 18 percent is grants, dominated by Medicaid. Mississippi's 2.57 is mostly old people's retirement and old people's health care arriving in a poor state with an old population, collected under a progressive income tax from states where the incomes are higher. It is not food stamps. It is the thing everybody says they support, working exactly as designed, and it flows uphill from the places that vote against it.

And the states with the highest poverty rates, the shortest life expectancies and the worst educational outcomes are not the socialist hellholes of California and New York. They are Mississippi, Louisiana, Alabama and West Virginia. The freedom those states offer is the freedom to be poor, sick and badly educated without government interference.

The claim. Extreme wealth is the reward for extreme contribution. The people who built these companies are the ones standing between you and a government that wants to run your life.

The record. Consolidation at the top is now large enough to buy the government outright, and one man has already run a piece of it.

The richest person in China is Zhang Yiming of ByteDance, worth roughly a hundred billion dollars on the 2026 Forbes list. The richest American, Elon Musk, was worth around $870 billion in August 2026. That is not the same order of magnitude. But the difference that matters is not the number. Zhong Shanshan, who held China’s top spot until this year, never ran a government agency. Musk did. He spent the first five months of 2025 running DOGE — a federal judge found in March 2025 that he had likely exercised authority only a properly appointed officer may exercise, though that was a preliminary ruling and was stayed within ten days — firing federal workers, cutting programmes and reshaping departments while SpaceX held some $16.7 billion in obligated federal contracts. He left at the end of May 2025 and told Katie Miller’s podcast that December that it had been only "somewhat successful" and that he would not do it again. The precedent is set. No Senate confirmation. No congressional oversight. No public account of who was even working for him.

As of late 2022 America’s 735 billionaires held $4.5 trillion between them, against $4.1 trillion for the entire bottom half of American households. In the 2024 federal cycle around a hundred billionaire families put $2.6 billion into politics, about one dollar in every six spent by all the candidates, parties and committees combined. A New York Times count published in March 2026 cast a wider net and found roughly three hundred billionaires and their immediate families giving about $3 billion, nineteen percent of every reported federal contribution. In 2008 that share was three tenths of one percent. They are not donating to democracy. They are purchasing it.

Nobody thinks the government has their best interest in mind. That is not naivety. That is precisely why you want systems with real checks on power, and right now we do not have them. The wealthy buy the legislation, they buy the elections, and they buy the judges who interpret the laws. Under this system that would happen even if it were illegal. It is legal. That is not a bug. That is the feature.

The displacement is real and I will not pretend otherwise. Writing, translation, paperwork, anything done at a screen at volume, goes first. Anyone who tells you it will all work out is guessing.

Here is what has actually been measured. Stanford’s Digital Economy Lab, working from payroll records covering millions of American workers, finds employment for twenty-two to twenty-five year olds in the most AI-exposed occupations sitting about nineteen percent below where it would be had it kept pace with less-exposed work, on June 2026 data. It shows up as reduced hiring rather than firing, and in headcount rather than in pay. Experienced workers in the same occupations do not show the same gap. On the freelance platforms the effect is measured and specific: writing work down about two percent in jobs and five percent in earnings after ChatGPT in one study, and postings for automation-prone writing and coding down twenty-one percent within eight months in another. In translation, each percentage point of machine-translation uptake tracks with seven tenths of a point less growth in translator employment, and roughly 28,000 translator jobs that were never created between 2010 and 2023. The Federal Reserve’s own working paper this March found coder employment growth decelerating sharply after 2022 — on the order of half a million jobs that would otherwise exist — while warning in the same paper against reading that as cause.

What is disputed is how much of it is the machines. The Bureau of Labor Statistics names automation, AI included, in its projections to 2035, where office and administrative support loses 752,100 jobs, the largest decline of any occupational group. But Yale’s Budget Lab, working from the national household survey rather than payroll records, finds no clear AI footprint in the labour market yet, and a Danish study with unusually clean identification rules out effects larger than two percent two years in. Nobody credible is currently claiming measured, economy-wide displacement. What is measured is young people in exposed work not getting hired.

Here is the rule I run my own life on. If a machine can do it, it gets cheap, so give it away or stop competing there. If it takes your hands, your presence, or your name on the line, it keeps a price. I sell tools I made at a bench from wood I cut. I give the writing away.

And look at what is already happening in Kansas. Farmers selling beef straight to families instead of through the packers. The middle is what dissolves. The two ends, the person who makes and the person who eats, are what remain. That is the tradesman’s world coming back. The family business. The man whose customers know his face. If you have a trade, you are better placed than the office worker who used to look down on you.

Somebody is driving your wages down. It is worth being precise about who.

When a company closes a plant in Ohio and reopens it in a country where labour costs a fifth as much, that is a decision made in a boardroom by people who will never meet you. When four firms end up owning every processor in a supply chain, the wage they offer is the wage, because there is nowhere else to go. Those are the mechanisms. The man standing next to you on the line did not design either one.

Here is what the numbers say about him.

The Cato Institute, a libertarian think tank, not a progressive one, has tracked criminal incarceration by immigration status for fifteen years, using Texas data, because Texas is the only state that records immigration status at arrest and checks it against federal records. In its March 2026 briefing paper, covering 2024, the incarceration rate per 100,000 adults aged 18 to 54 was 1,195 for native-born Americans, 674 for illegal immigrants, and 303 for legal immigrants. Illegal immigrants were about 44 percent less likely to be locked up than the people born here. Legal immigrants were about 75 percent less likely. The illegal-immigrant rate has risen since 2022, when it was 538, and it is still far below the native-born rate it has never once exceeded in the series.

On convictions rather than incarcerations, Cato's Texas figures for 2019 run 1,190 per 100,000 for native-born Americans, 749 for illegal immigrants, 510 for legal immigrants.

On the budget: Cato's February 2026 study of federal, state and local finances from 1994 to 2023 found immigrants reduced deficits by 14.5 trillion dollars over those thirty years. Immigrants supplied 14 percent of tax revenue while accounting for 7 percent of government spending. Illegal immigrants alone reduced the deficit by at least 1.7 trillion dollars.

On welfare: in 2023, non-citizens consumed 5,041 dollars per person in means-tested benefits and entitlements against 10,772 dollars for native-born Americans, 53 percent less. Non-citizens were 7.5 percent of the population and took 3.2 percent of the benefits. Naturalized citizens are the exception: they consume about 20 percent more than the native-born, and the categories where that shows up are Social Security and Medicare, which is what happens when a group is older than average.

Now the part that should decide the argument for anybody who cares about fairness.

In 2022, undocumented immigrants paid 96.7 billion dollars in taxes, 59.4 billion federal, 37.3 billion state and local. Of that, 33.9 billion dollars went into Social Security, Medicare and unemployment insurance: 25.7 billion, 6.4 billion and 1.8 billion respectively. Programs they are barred by statute from ever collecting from. Thirty-five cents of every dollar they paid went into a pot with their name struck off it.

The bar is not an accident and it is not vague. The 1996 welfare law makes undocumented immigrants ineligible for food stamps, non-emergency Medicaid, CHIP, TANF, SSI, Social Security retirement and disability, federal housing assistance, unemployment insurance, federal student aid, and coverage on the health insurance exchanges, including coverage they would pay for themselves, unsubsidised. The Earned Income Tax Credit requires a Social Security number; an ITIN does not qualify. Lawfully present immigrants who are not citizens generally wait five years after getting that status before most of those doors open at all.

They file anyway. In tax year 2022, nearly 3.8 million returns were filed using an Individual Taxpayer Identification Number, the number the IRS issues to people who have no Social Security number and want to pay their taxes regardless.

The effective state and local tax rate on undocumented immigrants that year was 8.9 percent. The rate on the top one percent of earners was 7.2 percent. In forty states, the undocumented paid a higher share of their income in state and local taxes than the wealthiest residents of that state.

On wages, the National Academies of Sciences reviewed the whole literature in 2017 and put the long-run effect of immigration on native-born wages at somewhere between minus 0.4 and plus 0.1 percent. That is a rounding error. Meanwhile, the CEO-to-worker pay ratio went from 20-to-1 to 281-to-1 and productivity outran compensation by nearly three to one. Nobody crossed a river to do that to you.

And on the line everybody uses: get in line, do it the right way. There are 50,000 diversity visas a year against more than 20 million applicants, and the lottery excludes Mexico, Guatemala, Honduras and El Salvador. Employer sponsorship is capped at roughly 130,000 a year for the whole country. For a working person without a degree there is no permanent line to stand in. You cannot tell somebody to use a door that was bricked up before they got there.

The claim. Benefits should be conditioned on work. Attach a requirement and the people who can work will, and the ones who will not should not be carried by the ones who do.

The record. It has been tried and measured. It does not produce work. It produces removals.

Arkansas put a work requirement on its Medicaid expansion in 2018. About 17,000 adults were taken off the rolls in three months. The uninsured rate among Arkansans in their thirties and forties went from 10.5 percent to 14.5 percent. Employment in the targeted group fell, and it fell in the comparison groups too, so the policy produced no employment gain at all. A third of the people it applied to had heard nothing about it. Nearly half were unsure whether it applied to them. That is the New England Journal of Medicine, 2019.

Food assistance has a longer record and it says the same thing. The best-identified study of it, published in the American Economic Journal in 2023, found that work requirements raised programme exits by 23 percentage points and cut participation among those subject to them by 53 percent, with — in the authors’ own words — no effects on employment. Homeless adults were screened out at disproportionate rates.

That is the policy now being expanded. The law signed in July 2025 puts an eighty-hour-a-month work requirement on the Medicaid expansion population across some forty-three states from 1 January 2027, and raises the food-assistance time limit to cover adults up to sixty-four while removing the exemptions that used to cover veterans, people with no address, and young people out of foster care. The Congressional Budget Office scores the Medicaid requirement alone at 5.3 million more people uninsured by 2034, and about 2.4 million fewer people on food assistance in an average month.

Now put the two halves of this page beside each other. The requirement points at entry-level, screen-based, clerical work. That is precisely the category where hiring is measurably falling, and where the government’s own projections expect three quarters of a million jobs to go by 2035 with automation named as a cause. A work requirement is a bet that the jobs are there. The evidence on the other page of the ledger is that they are going.

The claim. Hard choices are forced on us by real limits. Somebody has to go without, because there is not enough for everybody, and pretending otherwise is how you get shortages.

My answer. Make survival scarce enough and people will fight over scraps, and then you can point at the fighting as proof that people are animals. The capitalist creates the conditions that justify his philosophy, then calls the result human nature.

The name for what that produces is artificial scarcity. It is worth being honest that the phrase belongs to political ecology rather than to economics — Lyla Mehta’s work on water in western India is where it is used most carefully, distinguishing scarcity as it is lived from scarcity manufactured through political processes to suit the interests of powerful players. The serious economic version of the same point is Amartya Sen’s, from 1981: famines happen without the food running out. What fails is people’s ability to command food, not the supply of it.

Here are the cases where both sides of the ledger are actually measured.

Insulin. A peer-reviewed estimate in JAMA Network Open in 2024 put the sustainable cost-based price of a month of insulin at between roughly two and eleven dollars, depending on the type. A ten-millilitre vial of Humalog lists at $66.40, and that is after Lilly cut its list prices by seventy percent in 2023. The Inflation Reduction Act caps what a Medicare patient pays at $35 a month. There is still no federal cap for anyone on commercial insurance or on none at all. Two bills to fix that were introduced this spring and neither has passed. The molecule is a century old.

Doctors. The Balanced Budget Act of 1997 froze the number of Medicare-supported residency positions at their 1996 level, and Congress has added roughly 1,200 in the thirty years since. The Association of American Medical Colleges projects a shortfall of up to 86,000 physicians by 2036. The limit on how many doctors this country trains is a spending cap from 1997. It is not a shortage of people who want to be doctors.

Food. In 2024 the United States generated about 70 million tons of surplus food, roughly 29 percent of the food supply, worth some $380 billion. About 1.7 million tons of it was donated — just under 13 percent of what could have been. In the same year 13.7 percent of American households, 18.3 million of them, were food insecure. That is the last year anyone will be able to make that comparison: the Department of Agriculture cancelled the survey in September 2025, ending thirty years of consistent measurement.

Housing is the case people reach for first and it is the weakest of the four, so here it is with the concession attached. There were about 4.75 million housing units actively for rent or for sale in the middle of 2026, against 745,652 people homeless on a single night in January 2025. But a housing market needs vacancy the way an economy needs some unemployment, empty houses in Baltimore do not shelter anybody in Portland, and a good share of what gets counted as vacant is wrecked, seasonal, or tied up in probate. The number is real. It does not do the work people want it to do. The stronger version of the housing argument is the one mainstream economists already make: Glaeser and Gyourko call the gap between what a house sells for and what it costs to build a regulatory tax, and find it higher in many places than any plausible harm from building.

The honest objection to all of this is worth stating, because it is a real one, and it is Kenneth Arrow’s. Charging more for a thing than it costs to make a copy of is often exactly what pays for making the first one. Somebody has to fund the research. So the argument cannot be that price above cost is inherently illegitimate. It has to be that in a particular case the exclusion has outrun any reason for it — and in each of the cases above you can say why. A hundred-year-old molecule. A residency cap from 1997. Food already grown and already paid for, going into the ground.

Look at where the money actually goes.

In fiscal year 2025, SNAP served an average of 42.4 million people a month, about 12.3 percent of everybody in the country, at a cost of 95.8 billion dollars in benefits. The average benefit was 188 dollars a month. Six dollars and change a day.

The state with the highest participation rate is New Mexico, at 21.5 percent. Then Oregon at 18.1. Then Louisiana at 17.5, Oklahoma at 16.9, West Virginia at 15.5, Nevada at 15.2, Massachusetts at 15.1, Pennsylvania at 15.0, New York at 14.9 and Illinois at 14.8. Those are May 2025 rates.

That is not a map of big cities. It is a map of poor states. The thing that predicts a high SNAP rate is not how many people live stacked on top of each other; it is how little the work pays.

And look at who is in the households. Seventy-nine percent of SNAP households contain a child, an elderly person, or somebody with a disability. Those households are 88 percent of all participants and take 83 percent of the benefits. Twenty-eight percent of SNAP households have earned income coming in, somebody in them is working, and among households with children it is 54 percent. The picture of a program for people who will not work does not survive contact with its own numbers.

Then there is what has happened to it in the last year, which almost nobody has noticed.

The 2025 budget law raised the age at which work requirements bite from 54 to 64. It removed the exemptions that had covered homeless people, veterans, and young people ageing out of foster care. It narrowed the exemption for parents to those with a child under 14. It restricted the geographic waivers that had covered depressed areas to places with 10 percent unemployment. The Congressional Budget Office scored it at 186 billion dollars out of the program through 2034, the largest cut in its history, and estimated more than two million people losing benefits in a typical month, including about 270,000 veterans, homeless people and former foster youth.

Then, in November 2025, for the first time since the program began, the benefits did not arrive. Two federal judges ruled the withholding unlawful on the same day. Partial payments went out. Full payments were ordered, appealed, stayed, and ordered again. The shutdown ran 43 days.

Enrolment between May 2025 and May 2026 fell from 42.2 million to 36.5 million, down 13.5 percent, 5.7 million people in twelve months. Arizona fell 53.8 percent. Georgia fell 36.5 percent. Florida fell 22.2 percent. Louisiana fell 21.5 percent. Oklahoma fell 19.6 percent. Texas fell 16.6 percent. State officials in more than one of those places have described the drop as driven substantially by paperwork and verification burden rather than by anybody's circumstances improving.

The states that lost the most are not the states people picture when they picture food stamps. They are the ones that vote to cut them.

The claim. Take away the fear and nobody gets out of bed. The whole thing runs on need, and a country that removes need removes the reason anyone does anything.

The record. We have run that experiment three times and it did not go the way the fear says.

In Stockton, California, 125 people got 500 dollars a month for two years with no strings. In the first year the share working full time went from 28 percent to 40, against a rise from 32 to 37 in the group that got nothing. Finland gave 2,000 unemployed people 560 euros a month through 2017 and 2018 and found they worked slightly more, not less, and reported less stress, less depression and more trust in other people. The biggest, run by OpenResearch, gave a thousand Americans a thousand dollars a month for three years. They worked about an hour and twenty minutes less a week. They were also more likely to be looking for work, more likely to say the work had to mean something, and by the third year the Black participants were 26 percent more likely to have started a business, and the women 15 percent. People did not stop. They chose.

What decides whether a country full of automated work falls apart is not whether the machines arrive. It is who owns them and where the money goes.

And your own tradition has an answer to that which most people on your side have forgotten. Chesterton and Belloc argued a century ago that the problem was never the machines but ownership gathered into too few hands, and that the cure was property spread wide: a shop, a farm, a trade, a share, for as many households as possible. They called it distributism. A machine in a workshop you own is a tool. The same machine owned by a company you rent from is a boss.

That fits what I believe about why any of us is here, and I will say it as mine. The purpose of life is to tell all the stories, and yours is one of them. There is no bottom to that. The stories that can be lived, made and explored are endless, and a machine that handles the paperwork does not use them up. It hands them back. Nobody who has ever been handed a free week said there was nothing left to do with it.

The claim. The country you grew up in is being dismantled — your church, your town, your traditions, the way your people lived — by people who hold it in contempt.

My answer. Something is being dismantled, and it is worth being exact about what. It is not local culture. The church on the corner, the farm, the shop, the family with its own ways: that layer is not going anywhere, and parts of it are coming back. What is dying is the layer that grew on top of it and told it what to do. Traditionalist Christianity is dying, and so is every other traditionalist religion, and so is every hierarchical culture. I am not saying that to enjoy anybody’s loss. I am saying it because of which half goes. The part that is dying is the part that told you which questions you were allowed to ask. The part that stays is the part that shows up when the neighbour’s barn burns, and that part is not in any danger.

The numbers separate the two cleanly. Church affiliation fell for thirty years, from about ninety percent of American adults in the early nineties down to sixty-two, and then it stopped — Pew’s 2023-24 Religious Landscape Study, published February 2025, finds it level since 2019, with the unaffiliated plateaued too. Eighty-three percent still say they believe in God or a universal spirit. What kept falling is membership. In 2020, for the first time since Gallup began asking in 1937, fewer than half of Americans belonged to a church. Forty-seven percent. The hunger outlasted the institution. The faith is not what is dying. The building is, and so is the part of it that stood between a person and God.

And the political identity stapled on top of your local one is younger than you are. This page has already walked through where it came from: the Powell memorandum in 1971, the Heritage Foundation in 1973, Cato in 1977, the Federalist Society in 1982, and four decades of money spent making a worldview feel like something you inherited. My position, stated as mine: the cultures dying now are the ones that only exist because that machine paid for them. Your family is not that. Your town is not that. Both were here before any party was.

The opposite of dying is what I actually see coming. Artisan work, the hipster trades, blacksmithing, IPA and mead, knitting, crochet, and more and more of it — all coming back, and about to accelerate because of the machines rather than in spite of them. My position is that the inherited side of culture, the crafts and the old trades, is about to boom, and that what it expands is the sheer number of experiences one person can have in a life.

Some of that is already measured. The National Endowment for the Arts’ participation survey, which the Census Bureau fields and which reached nearly 41,000 adults, found the share of Americans doing leather, wood or metal work rose from 6.6 percent to 8.9 percent between 2017 and 2022 — about a third more people in five years. Enrolment at vocationally focused public two-year colleges reached 871,000 in spring 2025, up 19.4 percent across five years, and rose again this spring by 2.8 percent, another twenty-four thousand students. Registered apprenticeships stood at roughly 680,000 in the 2024 fiscal year, up 114 percent on the decade before. And there were about sixty commercial meaderies in this country in 2003; there were more than five hundred in 2025.

So when somebody tells you your culture is being erased, ask them which layer they mean. The one your grandmother actually built is not the one in danger. The one that goes is the one that was sold to you.

The claim. Greed and competition are how people are built. Any system that assumes otherwise eventually runs into the wall of what we actually are.

My answer. The capitalist sees humans as fundamentally selfish, competitive and motivated by fear. The socialist sees humans as fundamentally cooperative, social and motivated by belonging. Both are right — about themselves.

The capitalist projects his own psychology onto everyone else. He cannot imagine working without the threat of poverty because he himself would not. He cannot imagine sharing without compulsion because he himself would not. He looks at social programmes and sees freeloaders because, if the roles were reversed, he would freeload. He builds systems designed for the worst version of himself, and then acts surprised when those systems produce the worst version of everyone.

The socialist builds his philosophy on what people do when they are not forced into artificial competition. Both are projecting. But one projection creates the conditions that make the other one look true.

Ronald Wright, attributing the thought to Steinbeck, wrote in A Short History of Progress in 2004 that socialism never took root in America because the poor see themselves not as an exploited proletariat but as temporarily embarrassed millionaires. Whether or not Americans actually believe they will be rich someday, they have been trained to identify with the rich rather than with each other. The billionaire is not your friend. He is not on your team. He would replace you with a robot tomorrow if it saved him a nickel, and the only reason he has not is that the robot is not ready yet — and that is the part of this page changing fastest.

The World Happiness Report’s 2025 edition found that of the fifty-nine countries it studied, the United States had the largest increase in deaths of despair — suicide, alcohol, overdose — between 2000 and 2019. Three quarters of those countries moved the other way. The capitalist promised that the fear of falling would motivate greatness. What it produced is a country too exhausted and too frightened to do anything but cling to what little it has, while a dozen men buy rockets and elections.

Here are the ten American cities over 130,000 people with the highest homicide rates in calendar year 2025, per 100,000 residents:

Jackson, Mississippi, 53.0. St. Louis, Missouri, 49.7. Louisville, Kentucky, 47.1. Birmingham, Alabama, 46.3. Memphis, Tennessee, 38.0. Baton Rouge, Louisiana, 36.4. New Orleans, Louisiana, 33.2. Detroit, Michigan, 32.7. Kansas City, Missouri, 27.0. Cleveland, Ohio, 25.4.

Every one of those ten cities is in a state that voted Republican in 2024. Jackson has been first in the country for five years running.

Baltimore is not on that list. Baltimore recorded 133 homicides in 2025, down 31 percent in a single year, the fewest in nearly half a century. Washington DC is not on it either: its first-half homicide rate fell from 17.5 per 100,000 in 2023 to 6.2 in 2026, a 65 percent drop.

Now the part that gets left out of the argument entirely.

The most rural counties in America have a firearm death rate 37 percent higher than the most urban ones. Gun suicide in those counties runs 76 percent higher. Gun homicide runs 46 percent lower. The researchers who published that in JAMA Surgery in 2023 put the finding in one sentence: the risk of gun suicide in the most rural American counties exceeds the risk of gun homicide in the most urban ones.

In 2024 there were 44,447 gun deaths in the United States. Sixty-two percent of them were suicides. Thirty-five percent were homicides. The states with the highest total gun death rates were Mississippi at 28.0 per 100,000, New Mexico at 26.6, Alaska at 24.4, Alabama at 23.7 and Wyoming at 23.4. Wyoming and Alaska have among the lowest murder rates in the country. They are near the top of that list on suicide alone.

Counting every kind of injury death, accidents included: the rate in rural America is 31.6 per 100,000 against 24.0 in the cities. The more rural the place, the higher the chance of dying by accident or on purpose. That was published in the Annals of Emergency Medicine, drawing on 1.3 million deaths across all 3,141 US counties.

And the crisis everyone is arguing about ended four years ago. Murder fell 14.9 percent in 2024, 18.1 percent in 2025, and 17 to 18 percent again through the first half of 2026. The FBI describes 2025 as carrying the largest year-to-year decline in violent crime since it began estimating in 1936. The murder rate in 2025 was 4.1 per 100,000. The Council on Criminal Justice's assessment is that the rate is likely now the lowest recorded in any American dataset going back to 1900.

You are being sold fear about a thing that is disappearing, by people who need you afraid, while the deaths that are actually happening, in your county, to men you know, with their own guns, alone, are not on the news at all.