Actually, $3.5 Trillion Is a Lot of Money

O n Wednesday, Eric Levitz of New York Magazine took a bold stance in defense of President Biden’s massive American Families Plan (AFP): “$3.5 Trillion Is Not a Lot of Money.”

$3.5 trillion comes out to $27,247.74 for every household in the United States. When you add to that $3.5 trillion the $550 billion in new infrastructure spending Democrats have also proposed and the $1.9 trillion they already spent on COVID relief in January, the per-household cost rises to $46,321.17. The Congressional Research Service calculated in 2010 that the entire cost of the Second World War, in fiscal-year 2011 dollars, was $4.1 trillion. Democrats want to enact $5.95 trillion in new spending in just this year.

Now, I give Levitz credit for not just hiding behind the pretense that Democrats are doing something routine. “If passed in its current form,” he writes, “the reconciliation package would be the largest expansion of the American welfare state in half a century.” His fundamental argument, however, is a double-edged sword. Levitz complains that the $3.5 trillion number is artificially made to sound huge by the fact that congressional budget rules require an estimate of a proposal’s cost over ten years, so that an annual layout of $350 billion turns into $3.5 trillion. That is true: Democrats are not proposing a one-time, immediate $3.5 trillion layout in 2021. But it also underlines what Democrats are trying to do here: With a 50–50 Senate, a 220–212 House majority, and an unpopular president who was elected to just not be Donald Trump, they want to enact a permanent expansion of the federal budget that no future Congress will be asked to reauthorize.

If this were an infrastructure plan, its defenders could at least point to the promise of tangible, concrete results: When you borrow money to build a bridge when interest rates are really low (as they are now), at the end of the day, you will have a bridge. But the AFP proposes to spend most of that $3.5 trillion on welfare benefits such as Medicaid expansion, which is more akin to taking out a second mortgage to give your kids an allowance.

As I have written before, permanent spending authorizations are not compatible with the way in which our constitutional system was designed to operate. Every new Congress was supposed to be able to use the simple power of each House to prevent things from happening by saying no, and the president was empowered to do the same with his veto power. In England, the legislative power of the purse had allowed Parliament to resist the king, and the Framers assumed that it would discipline the presidency in a similar manner. Turning that presumption on its head, so that one needs both Houses of Congress and the president to affirmatively repeal spending once it is embedded in the budget, has all sorts of distorting effects, including on the separation of powers and on how we compute the size of spending bills. That has been compounded by Congress’s recent habit of presenting the entire budget on a take-it-or-leave it basis each year, so that the only way for lawmakers to say no to a penny of spending is to shut down the government.

In fact, because the design of most of the American Families Plan is to be permanent, capping an estimate of its costs at ten years massively understates what it will actually end up costing us if enacted. The additions to baseline spending are intended to continue until the sun burns out and the Earth is a frozen rock. People whose forgotten ancestors are yet unborn will be saddled for their entire lifetimes with the cost of Democrats’ winning David Perdue’s Senate seat two months after he outpolled Jon Ossoff on Election Day, and they will have little way of changing that. Which is exactly the political situation that Levitz and the backers of the AFP intend.

In theory, Republicans should use the fact that the $3.5 trillion won’t all be spent in 2021 as an opportunity to draw a line in the sand, and promise to repeal every penny of future spending the instant they have the power to do so. And if we still operated under a system of government in which the mere withdrawal of consent by one newly elected chamber of Congress could accomplish that goal, drawing a line in the sand would be an entirely practical political tactic. If we still operated under such a system of government, House Republicans could have killed Obamacare in 2011, when everyone was still exercised about it and the whole program had not yet become entrenched. Instead, they had to wait six more years and still could not muster the political will to repeal it, at least in part because repeal required getting the House and the Senate on the same page.

The great bulk of the modern federal budget consists — no matter which party is in power — of paying for things authorized long ago during periods of united Democratic Party rule. That includes the debts we keep accruing because the American people do not want to pay the exorbitant taxes that would be needed to support the rates of autopilot spending we already have. This is why charts and graphs showing what gets spent under different parties are all misleading, and all misleading in the same direction. Our deliberative system of government is supposed to make it harder to do permanent things, precisely so that no single election results in irrevocable changes to American society. We enshrine things in permanent law for good reasons, but there are also good reasons that we require cumbersome deliberation before doing so. Federal spending was supposed to be a for-a-season thing, revisited afresh after each election. The distorting, anti-democratic, dead-hand nature of a presumption of permanence for the spending of a single Congress is a fiscal straitjacket. And as spending spirals, debt mounts, and interest rates inevitably rebound from their historic lows, we will sooner or later be paralyzed by that straitjacket.