As the end of the year creeps up, Congress’s to-do list grows longer. Between infrastructure, reconciliation, and the “regular” annual spending bills, trillions of federal dollars are (potentially) sloshing about. A seeming similar number of press releases touting the benefits, or bemoaning the detriments, from the spending are popping up as well. It seems appropriate to take a moment to assess exactly how much spending Washington is contemplating and how that affects taxpayers.

After sitting on the shelf for months, the Senate’s bipartisan infrastructure bill is at the President’s desk ready to be signed on Monday. While some bill backers tout it as transformative and detractors as destructive, they both are overselling it. To start, the price tag is not $1 trillion+, it’s barely more than half a trillion dollars. This is because many are counting both NEW spending plus spending that would have occurred anyway without the legislation. What in Washington is known as “baseline” spending.

While new climate resilience programs are getting many of the headlines, the infrastructure bill also contains a full reauthorization of the once-every-five-years Highway Bill. When the baseline spending from the Highway Bill (which also funds mass transit), power grid, port, broadband, electric vehicle, and other programs is excluded from the $1.2 trillion top line, the amount of new spending is approximately $580 billion. Even the White House’s original talking points pointed this out in the *notes. While half a trillion is nothing to sneeze at, in a $23 trillion annual economy, it’s not as big as it first appears.

The bill is, however, a prototypical example of Congressional budgeting. The bill is not paid for by any reasonable measure. That’s the official ruling from the Congressional Budget Office (CBO) — $256 billion net increase in deficits — and common sense. The greatest claim of savings, nearly $200 billion, comes from repurposing dollars from various “emergency” COVID-19 response programs that cost less than originally estimated. Repurposing unspent “emergency” funds is a gimmick that increases deficits. It’s as if you borrowed $2,000 to fix your car, but ended up only being charged $1,200, so you spent the $800 you “saved” on new clothes, a nice lunch out, and some fuzzy dice to hang from your rearview mirror.

The bill backers also claim savings for things that already happened ($67 billion from a c-band spectrum auction in February 2021), or that might never accrue ($10 billion in future spectrum sales, $6 billion from selling oil out of the Strategic Petroleum Reserve, and $28 billion from cryptocurrency tax reporting). In the end, revenue from these provisions, even if the actions happen, is speculative because they would not occur for years and are dependent on future economic conditions. This is also true of the second largest pay-for cited – $53 billion in “economic growth” resulting directly from the projects in the bill. The bill’s backers can defend their work as important, or vital to the nation’s interests, but they can’t claim it’s paid for.

The issue of costs, both real world and in Congressional scoring, are at the center of the other hot topic, reconciliation. Originally touted as a $3.5 trillion package, the price tag has come down to $2 trillion, then $1.5 trillion, and now maybe $1.75 trillion? It’s a bit of a black box because there is no final bill, so no official score. And CBO has stated it won’t have all the preliminary scores of various finished sections of the latest package released by the House Rules Committee out until around Thanksgiving. One thing that is clear is the current 2,135-page package relies on many of the same Congressional accounting measures as the Infrastructure Investment and Jobs Act (the aforementioned bipartisan infrastructure bill).

The White House and Democrats tout their reconciliation bill as paid for. In some respects this claim is on sounder footing as the nonpartisan and official arbiter of tax provisions, the Joint Committee on Taxation (JCT), finds one version of the bill would raise nearly $1 trillion in revenue. But this, like other scores, is over ten years. It also assumes the various tax increases and credit reductions survive the sausage making process this year and into the future. The White House estimate banks on $480 billion from closing the tax gap between what is owed and what is actually paid, a goal we support, but one in which future revenue can’t be guaranteed. Finally, the price tag is being “reduced” by temporarily extending costly, but popular (at least with some), programs for a few years. An enhanced child tax credit and an increase in the limit on the state and local tax deduction are all likely temporary in the bill, but perpetual in reality.

Alright, we know budget scores are (a bit) nerdy. But money matters. And given the reconciliation bill’s lofty goals on everything from climate change to paid leave, the numbers need to line up. Enter CBO – the nonpartisan congressional spending scorekeeper which fills the same role as JCT on taxes. We recognize that scores are not guarantees. They are projections that try to predict the behavior of people and businesses over a future decade when there inevitably will be changing economic conditions. Sure, there are ways to game the system as there are in any rules-based system, but it is important to have the scores to compare one policy to the next. That’s why CBO should periodically look back at their work and see what they got right and wrong and publish the results so the rules can be refined.

Lastly, even with some gamesmanship and funny numbers, it is worth noting that the White House and Democratic leaders are at least trying to pay for the package, something we recommended they do. Even though they are currently estimating the reconciliation bill be more than paid for, it will most likely come up short. But being short is still better than massive deficit-financed proposals that some have called for and Republicans pushed through in their deficit-financed 2017 tax cut.