THE LIMIT CASE · ECONOMIST ARC — PART 4 · FALLING ACTION
Stop taxing labor; tax energy. The policy stack that sits on the new monetary architecture — an energy tax at the wellhead, a citizen's dividend derived rather than chosen, and a debt-first order of operations that lets the budget itself decide when the country can afford the next step.
Every tax system we have was built for a world of scarcity — a world where output grew about one and a half percent per person per year, and the job of policy was to squeeze a little more from a slow engine. In that world, friction is the enemy. You lower it wherever you find it: cut the frictions on trade, on capital, on hiring, on production, and hope the extra half-point compounds.
The take-off inverts the problem. When the surplus is large — when capital and energy can do most of what labor used to — the binding constraint is no longer how to grow. It's how not to waste the growth on disruption, on displacement, on a distribution that leaves most people holding nothing. Friction stops being the drag on a slow engine and becomes the steering wheel of a fast one. The question Part 4 answers is: if you had to point that engine, where would you put the friction, and where would you take it away?
The answer is symmetric, and it is the whole of this piece. Put the friction on energy. Take it off labor.
For two centuries the tax base was labor, for a plain reason: labor was where the value was made. Income and payroll taxes rode on the back of the factor that did the producing. Tax the thing that does the work — that is the durable instinct, and it was right.
The take-off severs the two. Labor is the term that, in the limit, trends toward zero — a shrinking share of what gets produced traces back to human hands — and the thing displacing it is energy: capital running on energy now does the work labor used to. Taxing labor in that world is doubly perverse. It penalizes the very factor we want to protect, and it speeds the displacement, because an income tax raises the price of a worker relative to the machine that replaces her.
So the move isn't to invent a clever new levy. It is to follow the value: take the tax off labor and put it on what has taken labor's place. Energy is the new labor — the factor that actually does the producing now — and it should carry the load labor used to.
This is also the answer to an instinct that keeps surfacing: the call to tax the robots, or tax the AI, that takes the jobs. The instinct is right — whatever displaces the worker should carry the load the worker used to. But the handle is wrong. "A robot" and "an AI" are no more definable for a tax collector than corporate income is, and they reorganize away just as fast: is a self-checkout a robot? a spreadsheet macro? a model served from three countries at once and rewritten every week? Chase the definition and you are back to taxing corporate income, watching the base dissolve under the very pressure you are trying to tax. But underneath, every robot and every model is the same thing — concentrated capital running on energy. You do not need to define the robot to tax it. You tax the energy it draws, metered at the wellhead and the substation, and the definition problem disappears. Taxing the robots is taxing energy.
So the tax base moves from labor to energy. This is not a new tax stacked on the old ones. It is a replacement: the Social Security tax, the payroll tax, and the income tax all come off, in that order, and an energy tax takes their place. Friction moves from the factor we want to free to the factor we want to steer.
The tax is levied upstream — at the wellhead, the mine, the generator, and the import border — on the energy that enters the economy: about 88 quadrillion BTU a year. Because it is collected where energy enters rather than where it is burned, it propagates cleanly through every price downstream. The meter on your house keeps doing its old job for the utility; the government never needs it — no new meter, no return to file, nothing to report. The tax simply becomes part of what things cost, the way a fuel cost already is.
The rule that defines the base is one line: the taxable event is energy entering the economy, not energy moving within it. Three ways in, and only three — extraction of a combustible fuel, generation from a primary non-combustion source, and import. Everything else is untaxed by construction. That matters more than it sounds, because it disposes of a whole category of objections without a single carve-out. Grid storage — batteries, pumped hydro, gravity, hydrogen — is not a source; it is a smoothing function, infrastructure, and it never pays twice. Nobody has to write an exemption for hydrogen; hydrogen simply isn't a way in. A whitelist needs no exceptions, which is why it beats the alternative.
Two bases, by physics rather than by fiat. Combustible fuels are taxed on the primary energy they consume — including the sixty-odd percent lost as heat in conversion. Non-combustion sources are taxed on the electricity they actually deliver: hydro, solar, wind, and nuclear alike. A lossy carbon source therefore ends up roughly twice as expensive per usable kilowatt-hour as a clean one. Price does the sorting. Carbon fuel isn't banned — it's a bridge, and a finite one; the tax simply lets its demand fall as cleaner supply gets cheaper. A company that puts solar on its roof to dodge the tax is doing exactly what the policy wants.
Nuclear deserves a sentence, because the obvious rule gets it exactly wrong. Taxed on primary heat, nuclear would pay more than coal — its steam runs cooler, so its heat rate is worse, and a rule that meant to price carbon would have priced thermal efficiency instead and killed the cleanest firm power we have. The fix is to tax nuclear the way its physics deserves: the energy that matters is the energy spent concentrating the fuel, not the decay heat that would have happened in the ore regardless. So nuclear pays its production cost, and then the grid rate on what it delivers — slightly worse than wind or solar, nowhere near coal or gas.
The rate is not a switch; it is a published, gradual ramp — two cents a year, every year, for twenty years, to forty cents per kilowatt-hour, stated in today's energy. That is the same pace the original design asked for; only the finish line moved. Predictability is the point. A business can plan two decades of capital investment around a rate it can see coming; it cannot plan around a rate that could lurch overnight. And twenty years at that pace spans a full turnover of the car fleet and a furnace replacement cycle — households and firms adapt when the old equipment dies anyway, instead of scrapping working capital. The schedule is a promise, and the promise is what makes the adjustment cheap. What the endpoint actually means is a subtler question than it looks; §10 takes it up.
And the adjustment is cheap. US energy spending runs about $1.6 trillion a year, near six percent of GDP (EIA, 2023). A tax on that scale is roughly a six-fold increase in what the economy pays for energy. Passed fully into prices, that is a one-time rise in the general price level on the order of twenty percent — spread across a twenty-year ramp, about one percent a year, at or below ordinary inflation and known in advance. The real bite falls only on the genuinely energy-hungry — aluminum, steel, cement, aviation — which is where you want the pressure to build efficiency and clean supply.
It is fair to call this a consumption tax, a cousin of the value-added taxes that fund most of the developed world — which invites the skeptic's arithmetic. A European VAT at nineteen to twenty-five percent raises seven to eleven percent of GDP; how does a tax on a single input raise twenty-three? Two ways. First, this is not a percentage added at the register — it is a several-fold increase in the price of one input, collected on every unit of it that enters the economy. Second, the base is bigger than consumption: a VAT exempts investment, government purchases, and exports by design, and energy is embodied in all three. A VAT reaches what shoppers buy. This reaches everything the economy does.
An energy tax reaches capital through its physical footprint — a data center is concentrated energy, and taxing the energy taxes the capital. But a floor under concentrated wealth belongs in the design too, for fairness and for the fortunes that throw off little taxable income while compounding out of sight. So: two percent a year on net worth above $100 million, phasing in from fifty, applied to individuals and to corporations alike, net of liabilities. Two percent is deliberately below what fortunes that size reliably earn — diversified holdings at that scale compound well above it — so the tax does not shrink great wealth; it slows the rate at which it pulls away from everyone else's.
Two notes. First, a naked federal wealth tax is on shaky constitutional ground — it looks like a "direct tax" that the Constitution would require apportioning among the states by population, and the Supreme Court pointedly reserved the question in Moore in 2024. The workable path is the one the corporate minimum tax already walks: levy it as a minimum tax on book income, which is unambiguously an income tax, and pursue a constitutional amendment for the pure net-worth version as the long game. Second — and this matters for how the whole piece is read — the capital floor raises under a tenth of the total. This is not a wealth-tax-and-UBI scheme dressed up. It is an energy tax with a dividend; the wealth floor is there for fairness, not for funding. Saying so changes who will listen.
If we tax energy at home, imports made with cheap dirty energy abroad would undercut us. So the same rate applies at the border to the energy embodied in imported goods — not a negotiated tariff, just "the same rate we charge ourselves." A country that wants the lower clean rate proves its energy mix; a published rate card replaces a coordination treaty.
The symmetric half matters just as much: energy leaving is refunded. We tax consumption, not production — the molecule that gets exported was never consumed here, and taxing it anyway would price American energy exports out of existence while inviting retaliation and collecting nothing. Mechanically it is a refund on export, not a deferred collection: the tax is charged upstream once, and repaid when the goods leave. The distinction matters: a refund keeps the collection point simple and the audit trail short; a deferral would put a rebate-shaped hole at every point in the chain.
The refund is what keeps our exporters whole. Consider an aluminum smelter, or a plant making optical glass by soot deposition — burning fuel so the soot condenses into the product itself. Both are heavily energy-intensive, and both sell much of their output abroad. Under an origin basis, those plants are dead on the day the ramp starts. Under a destination basis they are untouched, and their European competitors are the ones paying our rate at their border. The intent is to prevent the tax from breaking the competitiveness of things we make and sell abroad. Whether the refund should be the full embodied amount or something short of it is a fair question for public debate; that it must exist is not.
This is also what answers the natural question — what if everyone did this? Nothing about the machinery changes, and that is the point. Even between two adopters, exports are still refunded and imports still charged — the adjustment is what keeps each country's tax a tax on its own consumers at its own rate. Europe's VATs have been harmonized for half a century and every member still border-adjusts against every other, routinely, in both directions; nobody calls it a trade war. Drop the adjustment between two countries with different rates and the tax quietly becomes a production tax — factories migrate to whichever regime is cheaper, and the yardstick breaks. What adoption changes is the politics, not the mechanics: between adopters the border charge is symmetric, mechanical, and contested by no one — a customs formality rather than a wall — and protectionist tariffs lose their remaining job. A trade agreement can streamline the paperwork. Only a deliberate one could remove the adjustment, and neither side should want it to.
The refund carries a known failure mode, and it belongs on the page rather than in a footnote. Export refunds are the worst fraud vector in the VAT world — carousel schemes, claiming refunds on exports that never happened, cost the EU on the order of €50 billion a year ◦ — and ours would refund an estimated quantity, embodied energy, which every exporter gains by overstating. The design's simplicity is real everywhere except the border, and the border is where its auditors will live. We do not have a complete answer yet.
And the usual fear — that capital simply leaves — runs into two walls here. The first is physical: fleeing the tax means finding gigawatt-scale power somewhere else, and there is no somewhere else with spare gigawatts. Grid access is the binding constraint on compute everywhere on earth, and any country with surplus power to sell will notice it can tax it too. The second is the border: the duty on embodied energy extends to imported inference — AI compute bought from abroad pays the same energy rate at the border it would have paid at a domestic wall. This is not a general internet tariff; email and web hosting cost rounding-error energy and cross free. It is a duty on the one imported service that is, physically, packaged energy. You can offshore a smelter, and the border card prices that; offshore the datacenter and the border card prices that too. Leaving buys nothing.
Because energy carries the revenue, the entire apparatus of taxing people's incomes can be switched off — Social Security tax, payroll tax, income tax, all of it. The government no longer needs to know what you earn.
That single fact dissolves an enormous amount of friction. No individual filing, no withholding, no April 15, no four-hundred-billion-dollar annual compliance industry, and none of the income surveillance that comes with it. It is the same move the dividend makes on the benefits side — universal eligibility needs no eligibility police — applied to the funding side: universal funding needs no income police.
And every friction it removes is a friction on work — but not in the way that phrase usually means. The payroll tax is a tollbooth on entering the formal economy. Self-employment tax runs 15.3 percent from the first dollar of profit — both halves of the payroll tax, self-administered in quarterly filings — and for the potter, the busker, or the woman with an Etsy shop and a day job, the paperwork alone is reason enough never to start. It is a system built for a 1940s factory floor, and what it taxes is the act of becoming legible. Take it away and the distinction between formal and informal work stops existing, because the books are about energy now. There is no "off the books" to be off of. The informal economy doesn't get formalized; it just becomes the economy.
That is what participation means here: the goal isn't to push people into jobs, and it isn't to excuse them from work either. It's to remove the penalty on small-scale participation — so that working for a corporation, crafting things to sell online, and playing live music at a pub on Fridays are all simply people being in the economy, and none of them requires permission or paperwork to count.
One levy stays, and the exception proves the rule. Workers' compensation keeps its premium, because that premium is experience-rated: it prices the cost of an unsafe workplace and rewards a safe one. Keep the labor charge that's an incentive; drop the ones that are merely revenue. Taken together, this is arguably the most pro-market, pro-entrepreneur tax code ever seriously proposed — and it happens to also end poverty.
Energy gets dramatically more expensive under this plan. The plan has to say what that does to a household before it says anything else, because a policy that makes heating a moral question has failed no matter what it raises.
The answer is a dividend, and the size of it is derived rather than chosen. Households burn energy directly — heating, cooling, hot water, the car — and that direct burn is 27.7 percent of the entire base. So 27.7 percent of the revenue is rebated, per capita, in cash. The average household is therefore made whole on its direct energy at every single point on the ramp, by construction, at any rate. That property is the load-bearing one: the rebate is a share of revenue, not a fixed sum, so it scales automatically with the rate. Nobody has to legislate a cost-of-living adjustment. The arithmetic does it.
It has to be cash, not a credit on the utility bill, and the difference is the whole incentive. A bill credit can only be spent on energy, so saving energy saves you nothing extra. Cash means the panel on your roof is a machine that turns sunlight into anything you want — groceries, a guitar, a plane ticket. Same dollars; opposite behavior. Nobody forces a good choice; the price does the arguing and the household keeps the winnings. That is the free market doing what it advertises: the household, not a program administrator, decides where the next dollar goes. The government sets one true price and takes its thumb off every other scale.
Who actually comes out ahead is then a single question with a clean answer: you are net positive if your energy use is below average. Not below average income — below average energy. Some of what falls out of that is intuitive and some isn't.
That last point disposes of the objection that looks strongest at first glance — that rooftop solar is an arbitrage for people rich enough to own roofs. It's the reverse. Solar is the escape hatch for the energy-heavy, and the people who can install it are precisely the people who are underwater. Renters can't install it and don't need to; they're already ahead. Capability and need line up. (It also means community solar, which exists today to give renters access to solar economics, has nothing left to do. The problem it solves stops existing.)
What the dividend pays is a share of the revenue, so it starts small and climbs with the rate — never lagging the tax, because being whole on direct energy is the construction, not a promise. Add the couple hundred a month the monetary reform of Part 3 throws off as a property of the money supply, and the combined floor runs about $370 a month at year four, $615 at year ten, and $1,110 at the twenty-year endpoint, per person — with the economy growing at just 1.2 percent, before the take-off adds anything. (Hold GDP perfectly flat as the pessimistic floor and the path is $360 / $570 / $920.) For the studio dweller the endpoint nets to about $640 a month of purchasing power after their own energy bill.
Be precise about what that is. It is not a life. It doesn't cover a city studio's rent by itself. It makes that life viable with modest participation — the pub gig plus the floor — which is exactly the point and should not be oversold into something it isn't. The target remains a genuine basic income, on the order of twelve hundred dollars a month; §§9–10 price what it takes to get there.
Children get a full share. They are expensive, and a dividend that pretends otherwise is a dividend that punishes families.
But not all of it goes to the parents, and the split has a reason rather than a knob behind it. A child's own energy burn is about 69 percent of an adult's — they share the house evenly, but they don't commute. So seventy-five percent flows to the parent, which covers the child's energy footprint — the shared house, the heat, the miles. (Everything else a child costs is still on you, as it always was.) The remaining twenty-five percent is not assigned to the parent at all. It goes into a retirement account in the child's own name.
Nobody is docked. The child receives a hundred percent of their share: seventy-five now, twenty-five at sixty-seven. It is not a haircut, it is a maturity transform — and the numbers it produces are hard to believe until you run them. For a child born into the completed system — the year-twenty world, with the share at full size — eighteen years of contributions at a conservative four percent real come to about $53,000 at age eighteen. Left alone to sixty-seven, it is $362,000, and a four percent draw on that is $1,206 a month.
That is the twelve-hundred-a-month target, to the dollar. Nobody tuned it; it fell out. For scale: the median American household aged 65 to 74 today holds about $200,000 in retirement accounts ◦ — and that is the median among households that have accounts at all — while the average Social Security check runs about $2,070 a month ◦. An account nobody touched, seeded from a child's own share, arrives at retirement nearly double the median outcome of a lifetime of deliberate saving — and it arrives on top of the dividend, not instead of it. The generation that grows up under this arrives at retirement with a second basic income they started funding before they could walk.
The design should be familiar rather than exotic, because familiarity is doing real work here. It's a 401k. A handful of investment options with a broad index fund as the default; direction hands over to the owner at eighteen; they can add to it, including from their own dividend; the return is not guaranteed and a four percent draw is forced at retirement. Handing it over at eighteen is not a nicety — it is what makes the money the citizen's property, which is a far harder thing for a future Congress to raid than any promise or fiscal rule. Norway needed a spending rule to protect its fund. Property rights are stronger than spending rules.
The return isn't guaranteed, and that is acceptable because the floor lives somewhere else — the dividend and the monetary floor don't stop at sixty-seven, so the account is upside on top of a guaranteed base, not the base itself. Someone retiring into a 2008 watches $1,206 become $700 and still lands near $1,600 a month.
What else the account implies — a fund at the scale of the largest asset managers, a generation arriving at adulthood as owners of the country's capital — is a bigger story than a tax paper, and Part 5 tells it. Here it is enough that the children are accounted for, and that the dividend's cost is priced with them in it.
A plan like this lives or dies on sequence. If you cut labor taxes exactly as fast as the energy tax comes in, you have done something revenue-neutral — you've held the existing deficit in place and paid down nothing. Balance is not progress.
So the sequence opens with the thing nobody campaigns on: the debt. For the first years, nothing comes off. The energy tax climbs at its steady two cents a year while every existing tax stays in place, and all of the new revenue goes against the deficit. The first three years still run red — the ramp starts from zero — but around year four the budget crosses into surplus for the first time in decades, and the surplus then grows by nearly half a trillion dollars a year as the ramp climbs.
From there the reforms fire in order, and none of them fires on a date. Each waits for the same gate: a surplus of two trillion dollars — enough that the reform can take its bite and still leave a trillion a year hitting the debt. Nobody has to forecast the take-off, and nobody has to be right about the future in year one. The budget itself announces when the country can afford the next step. The gate is also the safety: the modeling is unambiguous that retiring labor taxes ahead of coverage is the one move that detonates the plan — done early against a slow ramp, debt passes two hundred percent of GDP. The gate makes that mistake structurally impossible rather than merely inadvisable.
Social Security goes first, when the gate first opens — year nine on the central path. It cannot go sooner, for a reason of arithmetic: the program roughly self-funds today — the payroll tax coming in pays the retirees drawing out — so ending the tax opens a $1.35-trillion-a-year hole that the energy ramp must first grow into. End it at year zero and the deficit jumps by seventy percent in exactly the years the sprint is supposed to be closing it. And the tax and the accrual must end together. Social Security's political armor is the conviction that the tax earns the benefit — the same conviction that has defeated every attempt to means-test the program also forbids collecting its tax while crediting nothing. So at the gate, both happen at once: the tax stops, and accruals freeze.
The freeze is a closed-group plan termination, a mechanism the private sector has run thousands of times winding down defined-benefit pensions. Every benefit locks in as if its owner never contributed again — which leaves anyone near retirement untouched, since they already have their thirty-five years, and gives a twenty-five-year-old about nine percent of a full benefit, plus the dividend for life and a retirement account newly open to their own contributions — not the funded account at birth; they were born too early for that. No cohort carries more of the bridge than theirs. The energy tax carries the closed group until the last recipient is gone. As they go, the money doesn't return to the Treasury; it flows into the dividend and raises the floor for everyone.
The instinct is that this must be expensive — two systems at once, across a sixty-seven-year bridge. It isn't, provided the books are read carefully. On the benefits side, the frozen plan never pays out more than the status quo in any single year: the freeze writes no new promises, and both worlds owe the same wave of retirees already earned. The revenue side — the $1.35 trillion the ended tax no longer collects — is real money, and it is precisely the hole the surplus gate exists to fill before the freeze fires. By year twenty-five the frozen plan pays out about $320 billion a year less; by year forty, $1.2 trillion; by year eighty it is zero, while the status quo is still running and always will be. Both face the same retirement wave — the status quo simply keeps writing new promises on top of it. The freeze pays for the wave and stops. The bridge isn't a cost. It's the runoff of a debt we already owe, and we would have paid more of it without the freeze.
The income tax goes next, from the bottom up. Not by cutting the low rates — a lower bottom bracket hands the same dollars to high earners, whose income passes through those brackets too — but by raising the threshold beneath which you owe nothing and file nothing. Every year the floor rises, millions more Americans never file again. The simplification §6 promises arrives progressively, household by household, rather than all at once at the end.
Two pieces stay standing on purpose: the corporate income tax and the top individual bracket hold until the net-worth amendment of §4 is ratified. They are the machinery the interim minimum tax runs on — and they are the incentive. The people best positioned to fund a fight against the amendment are the people still paying the old taxes. Ratify it, and the income tax dies in exchange for two percent on fortunes above a hundred million.
Universal health care comes at the gate after that — decoupled from employment, lifting its cost off every employer and every job, with the Medicare payroll tax retiring alongside it. And the dividend runs from day one — not as a reward at the end of the sequence, but because it is the thing that makes the ramp survivable, per §7.
On the central path — the economy growing at the modest 1.2 percent a year the last fifty years' fit suggests, spending growing right along with it, and health care arriving whole in its gate year, not on an installment plan — the cascade completes with room to spare: Social Security at year nine, the income taper from year eleven, universal health care at year sixteen, and the debt — 125 percent of GDP when the ramp starts — gone entirely at year twenty. Hold GDP perfectly flat instead, as a pessimistic floor, and every reform still fires within a year of the same schedule and the debt clears at year twenty-two. Let electrification erode the base a percent a year on top and it clears at year twenty-five. The sequence is robust; only its speed moves. One mechanism deserves its own sentence: the growth case is not optimism about energy abstinence reversing — the border duty prices the energy embodied in imports, so the offshoring that flattened America's meters for twenty-five years never left the tax base. The economy's growth reaches the revenue whether the factory sits in Ohio or arrives in a container.
And the cascade has a final gate: the year after the debt dies, the surplus that was paying for the past starts paying people. It folds into the dividend, which jumps from about $1,110 to roughly $2,240 a month including the floor — more than today's average Social Security check ◦ — in year twenty-one. Through the twelve-hundred-dollar target without slowing down. (On the flat-GDP floor the jump comes at year twenty-three, to about $1,690.) The sequence that began by refusing to give anything away ends by giving away everything it was carrying.
The sprint has a real cost: for the first several years people pay both the old taxes and a rising energy tax before the relief arrives. That is the price of paying the debt down rather than inflating it away. Inflating it away is a lever that exists; it is not one to reach for first.
One number needs defusing. Gross federal flows under this design run near thirty-eight percent of GDP, which sounds like a state that has doubled. It hasn't. About eight and a half points of that is the dividend — money that lands in citizens' accounts and never funds a government function. Net it out and the state runs at roughly twenty-nine percent of GDP, against about twenty-eight today. Apples to apples, the government does not grow. It is funded completely differently, and it does considerably more. We should print both numbers rather than the flattering one.
Here is the thing this piece has been circling. Forty cents a kilowatt-hour is not a number we chose. It's a bet.
The base shrinks as the policy succeeds — and not merely because people use less. Three separate mechanisms, all of them the design working as intended. Cleaning the grid moves generation from the primary basis to the delivered basis, cutting the taxed base for the identical kilowatt-hour. Electrifying transport delivers the same miles on about sixty percent less primary energy. Heat pumps deliver the same warmth on about half. Put them together and a fully clean, fully electrified America needs roughly 35 percent less taxed energy for the same life.
Revenue is rate times base. If the base falls by a third, the rate must rise by half just to stand still. At flat energy services the honest endpoint isn't forty cents — it's fifty-six. If services grow by a quarter, forty-five. If the take-off doubles them, twenty-eight.
So the forty-cent figure encodes a specific claim: that energy services grow forty to fifty percent over twenty years. We should say that out loud rather than present a number and hope nobody differentiates it.
Two things follow, and they cut in opposite directions, which is why both belong on the page.
The first is that a rising rate is not a failure. It is the scoreboard. Because the dividend is a share of revenue, it rises exactly as fast as the rate does. Same bill, half the kilowatt-hours, double the price. The household is indifferent by construction — and the household below average does better as the rate climbs, not worse. Our studio dweller nets $420 a month at forty cents and $610 at fifty-six. The endpoint going up is good news for the person this policy exists to protect. It is the strongest argument in the design — and it works only if the endpoint we quote is the one we actually expect, not the one that sells.
The second cuts the other way, and we should get there first. The arithmetic depends on the growth it exists to steer. Under flat GDP with a real efficiency response, energy nets the government under four trillion against a requirement near seven — it doesn't even replace the labor taxes it's meant to retire, and the debt doesn't fall. The plan needs the take-off. It is not a hedge against stagnation; it is a steering wheel, and a steering wheel is no use on a car that isn't moving.
There is a reason to think the bet is good, and it is the border card. American energy use has been flat for twenty-five years while output grew seventy percent — which reads like an efficiency miracle and mostly wasn't. We offshored the energy. We shipped heavy industry abroad and imported the goods back, and the intensity went with it. A rate card on embodied imports plus a refund on exports reverses exactly that: it makes energy-intensive production competitive here again. The border adjustment is a re-shoring engine, and re-shoring is base growth — a force the last twenty-five years of data cannot see, because it did not exist.
Three forces act on the base: the structural shrink, which is physics; re-shoring, which the policy creates; and the take-off, which is the thesis. Two of the three are ours. That is a bet worth taking. It is still a bet.
Strip it to the frame. Stop taxing labor; tax energy. Abolish the Social Security tax, the payroll tax, the income tax, the employer health mandate, and in time the minimum wage. Replace the welfare bureaucracy with a single dividend nobody has to qualify for. Pay down the debt, honor every pension already promised, cover everyone's health, put a rising floor under every citizen, and hand every child a retirement account on the day they are born — funded by the one factor that can't run away, steered by a price that sorts clean from dirty on its own.
None of the dials here are sacred, and that is a feature. The exact rate, the ramp speed, the dividend's climb — set them where the evidence points; the structure holds wherever you set them, because it changes what is taxed, not merely how much.
What's left is Part 5's job: to show that these pieces — the diagnosis, the factor shift, the monetary peg, and this tax stack — are not four policies but one operating system. And to answer the question this piece keeps brushing against without ever quite asking: what a country is for, once the energy does the work, and every citizen owns a piece of the thing that does it.