The Carbon Dividend

The problem

Almost every economist who has looked at emissions has reached the same conclusion, across a remarkably wide political range: the cheapest way to reduce carbon is to price it, and let a hundred million households and firms find their own abatement rather than having Washington pick it for them.

Almost every attempt to enact one has failed.

Washington State put a carbon fee on the ballot twice and voters rejected it twice. France introduced a fuel tax increase and got the gilets jaunes. Australia enacted a carbon price and repealed it within two years. The pattern is consistent enough to be a law of political physics: the cost of a carbon price is immediate, visible, and lands on the pump; the benefit is distant, invisible, and lands on everyone.

That is a losing trade for any household asked to accept it, and voters are not being irrational when they decline. They are correctly identifying that they have been asked to pay now for a diffuse benefit later, with no assurance the money will not simply be absorbed.

Meanwhile the design problem gets harder from outside. The European Union's carbon border adjustment mechanism moves to its definitive phase, which means American exporters of steel, aluminium, cement and fertiliser will begin paying a carbon charge to Brussels on goods made in Ohio. If America does not price its own carbon, its trading partners will price it for us and keep the money.

So the question is not whether American carbon gets priced. It is whether Americans collect the proceeds.

The proposal

A rising fee on carbon at the point it enters the economy, returned overwhelmingly to citizens as an annual per-capita cheque, with a minority share deposited as principal in the American Permanent Fund.

How it would work

1. Upstream, and small in number. The fee attaches at the wellhead, the mine mouth, and the port of entry, on the carbon content of fossil fuels. A few thousand collection points rather than a hundred million. This is the design that makes a carbon price administratively trivial compared with almost any regulatory alternative.

1. A published path. Start at $40 per ton of CO2 and rise on a fixed, legislated schedule of several percent above inflation each year. The schedule matters more than the level, because capital investment responds to the price in 2040, not the price today.

1. Most of it comes straight back. 75 percent of gross receipts are returned as an equal per-capita annual payment to every citizen, children at a half share, paid the same week as the Dividend Day of № 017. Equal shares, not income-tested, for the same reason Alaska pays every resident the same: a universal payment is understood as a share, and a targeted one is understood as a benefit, and only the first survives a change of administration.

1. A quarter compounds. 25 percent goes to the Fund as principal under the locks of № 023, so that the transition builds a permanent asset and not only a rebate.

1. A border adjustment, both ways. A charge on the embedded carbon of imports from countries without a comparable price, and a rebate on exports. This is what prevents the fee from simply relocating American industry, and it is what converts the EU's mechanism from a tax on American exporters into a credit against it.

1. It replaces things. Where the fee covers a sector, redundant regulation on the same emissions is retired. A price and a mandate aimed at the same molecule is two instruments doing one job at twice the cost.

The numbers

US energy-related CO2 emissions run around 4.8 billion tons a year.

A $40 per ton fee on that base is roughly $190 billion a year at the outset, declining per ton as the base shrinks and rising in total as the rate climbs.

The 75 percent rebate is about $145 billion a year, which across roughly 335 million people at a half share for minors is on the order of $500 per adult per year, before the schedule starts rising. A family of four collects about $1,500.

Against that, the fee raises household energy costs. The distributional work on fee-and-dividend designs consistently finds the same shape: because wealthier households consume far more carbon in absolute terms while the cheque is equal, roughly the bottom two thirds of households come out ahead in cash terms, and the poorest come out furthest ahead. That is not a rhetorical claim, it is arithmetic about who burns more fuel.

The 25 percent to the Fund is about $48 billion a year, the largest single stream this catalog has proposed. Compounded at 5 percent real for a decade it is roughly $630 billion of corpus, throwing off $31 billion a year, permanently, or about $95 per citizen per year on its own.

Now the part we have to say out loud, because it is a departure.

Every other dividend in this catalog sends all of its receipts to principal and pays nothing for years. This one sends three-quarters straight back to households immediately. The difference is who pays. A levy on compute (№ 031) or hardrock ore (№ 037) is collected from firms extracting value from a public asset. A carbon fee is collected, in the end, from a family filling a tank. Money taken from households in the present must return to households in the present. A carbon price that told working Americans to wait until the 2040s for their share would deserve to lose, and would.

The honest objections

"This has lost every time it has been put to voters. You have not solved that." The best objection, and we cannot claim to have solved it. What we can say is that the versions that lost were, almost without exception, not dividend versions. Washington's 2018 measure directed revenue to a state-appointed investment board. France's was straight revenue. Australia's compensation was routed through the tax code, where nobody could see it. The proposition that a large, visible, equal cheque with your name on it changes the politics is untested at national scale, and Alaska is the only real evidence that a universal resource dividend becomes politically untouchable. It is a hypothesis. It is the only one on offer that takes the political failure seriously rather than treating it as a communications problem.

"Returning the money defeats the purpose. People will just spend it on gas." This confuses the price signal with the income effect, and they are separate. The rebate is a lump sum, unrelated to how much fuel any individual burns, so the marginal cost of the next gallon is still higher for everyone. A household that cuts its consumption keeps more of the cheque. That is exactly the incentive you want, and it is why fee-and-dividend reduces emissions at close to the same rate as a fee that keeps the money.

"It is regressive." The fee is regressive; the fee-plus-dividend is progressive. That distinction is the entire architecture and it is not a rhetorical trick. But two real caveats survive it. Rural households, who drive further with fewer alternatives, do worse within any income band, and the flat cheque does not fully correct for that. And the transition is hardest on people in carbon-intensive employment, which is a different problem that № 026 addresses and this proposal does not.

"China emits more than we do. This is unilateral disarmament." China emits roughly twice what the United States does, and no American policy fixes that. Two responses. The border adjustment in item 5 is precisely the instrument that makes the price partly extraterritorial: goods entering the American market pay it regardless of where they were made, which reaches a meaningful share of the emissions in question. And the argument proves too much, because it is an argument against any domestic action of any kind, including the regulatory approaches its usual proponents prefer.

"Border adjustments start trade wars and may violate WTO rules." A serious legal question, genuinely unsettled, and the design has to be scrupulously non-discriminatory to have a defence under GATT Article XX. It is also a question the EU has already forced onto the table. America can help write the rules for carbon border measures or can be a rule-taker, and this is one of the few areas where being early is worth more than being cautious.

"$500 a year does not compensate me for a permanently higher cost of living." For most households the arithmetic says otherwise, and for some it does not, and the honest answer is that this proposal makes a majority better off in cash and a minority worse off. We would rather publish that sentence than the one where everybody wins.

Sources

• US energy-related CO2 emissions of approximately 4.8 billion metric tons; Energy Information Administration (eia.gov)
• Climate Leadership Council carbon fee-and-dividend framework; Baker-Shultz plan (clcouncil.org)
• Distributional analyses of carbon fee-and-dividend finding net gains for lower- and middle-income households; US Treasury Office of Tax Analysis working papers
• EU Carbon Border Adjustment Mechanism, transitional phase from 2023, definitive phase from 2026 (ec.europa.eu)
• Washington State Initiative 732 (2016) and Initiative 1631 (2018), both defeated; Australia's carbon pricing mechanism, enacted 2012 and repealed 2014; British Columbia's revenue-neutral carbon tax, 2008
• Alaska Permanent Fund Dividend as the precedent for equal per-capita resource payments
• Proposals № 017 (Dividend Day); № 023 (The Debt Covenant); № 026 (The Landing); № 031 (The Machine Dividend); № 037 (The Hardrock Royalty)