On Jan. 10, 2025, the Bureau of Labor Statistics released the last full employment report of Joe Biden’s presidency. Payrolls had grown by 256,000 in December. Unemployment stood at 4.1%.
Five days later the inflation number landed. Consumer prices had risen 2.9% over the previous 12 months.
Five days after that, Donald Trump took the oath of office.
Those three dates get cited constantly, almost always in service of an argument they don’t actually support. One side reads them as proof that Biden fixed the economy. The other reads the price level, which was still high, and calls the whole thing a wreck. Both readings skip the part that matters, which is what happened to the money afterward. Trump inherited relatively favorable headline conditions alongside real, documented weaknesses. Six months later, his administration and its congressional allies enacted a reconciliation law, and the Congressional Budget Office found that the law would leave households at the bottom of the income distribution with fewer resources while households in the middle and toward the top ended up with more. What changed hands in January was a set of conditions. What came after was a choice.
What the December numbers said
Unemployment at 4.1% is not a crisis reading. Neither is 256,000 jobs in a month. The pattern held through the transition: January unemployment came in at 4.0%, with 143,000 jobs added, in a report BLS published Feb. 7 covering a month that belonged mostly to the outgoing administration.
So the labor market Trump walked into was not collapsing. That’s the strongest single fact in this argument, and it’s worth stating flatly rather than dressing up. There was no mass unemployment. There was no employment freefall.
Inflation is more complicated. At 2.9% year over year, prices were rising faster than the Federal Reserve’s 2% longer-run objective. The figure was also a long way down from June 2022, when the all-items index hit 9.1%, the fastest 12-month increase since November 1981. Biden did not eliminate inflation. Nobody serious claims he did. He left office with it much closer to normal than it had been in the middle of his term.
There’s a distinction buried in that sentence that gets lost constantly. Inflation is the rate at which prices climb. When the rate falls, prices don’t come back down. They climb more slowly from a higher base. A shopper comparing a 2025 receipt to a 2019 receipt is measuring the base, not the rate, and the base had moved a great deal.
Now the weaknesses, because the argument needs them.
Housing was brutal. The median existing-home price in December 2024 was $404,400, the 18th straight month of year-over-year increases, and Freddie Mac had the 30-year fixed mortgage at 6.96% by late January. Existing-home sales for the full year came in at 4.06 million, the lowest annual figure since 1995. Interest rates were high across the board; the Fed had cut in December but left the federal funds target at 4.25% to 4.50%. Food, insurance and health care all cost substantially more than they had in 2019. Federal deficits were enormous and structural. And the military, despite its budget, had readiness and procurement problems that predate everyone currently arguing about them.
Anyone describing the economy Biden handed over as good, without qualification, is going to lose the argument to a grocery receipt.
Conditions are not policy
An administration can inherit low unemployment, falling inflation and rising wages without having created any of it. It can also inherit those conditions and then improve them, damage them, or leave them alone.
Credit for the starting position and responsibility for what follows are separate questions. Treating them as one question is how these debates end in a stalemate about who really caused what in 2021.
Set the credit question aside. It isn’t answerable in a way that will satisfy anybody. The answerable question is what conditions existed in January 2025 and what federal policy then did to the distribution of resources.
Well funded is not the same as ready
The claim that Biden left Trump a fully prepared military is not defensible, and it doesn’t need to be.
What he left was an enormously funded force with substantial global capabilities and a set of longstanding structural problems that funding alone hasn’t solved. The Government Accountability Office testified in March 2025 that U.S. military readiness had degraded over roughly two decades, that every service had struggled to field enough personnel, aircraft and ships capable of performing their missions, and that the Defense Department had yet to act on more than 100 GAO recommendations aimed at fixing it.
The problems GAO identifies are not primarily budgetary. They involve maintenance backlogs, acquisition delays, personnel shortfalls, modernization gaps and weapons programs that slip their schedules. In its 23rd annual weapons assessment, published in June 2025, GAO reported that DOD plans to spend nearly $2.4 trillion developing and buying its 106 costliest weapon programs, and that the average time for a major program to deliver even an initial capability had reached almost 12 years from program start. That average had grown by 18 months in a single year.
Twelve years. A program that starts when a kid enters kindergarten delivers its first working capability around the time that kid starts driving. GAO’s file is full of individual examples, including the Air Force’s B-52 radar modernization effort, which reported an additional 34-month delay tied to testing, parts and software. The Pentagon’s acquisition bureaucracy has been the subject of reform campaigns since roughly the Eisenhower administration without much to show for it. But the number is the number.
The useful formulation is that well funded and fully ready are different conditions. Keeping them separate protects the argument, because anyone who wants to knock down “Biden left a prepared military” only has to point at a GAO report to do it.
Public Law 119-21
The centerpiece of the second Trump term’s economic record is a reconciliation law.
The Senate passed it 51-50 on July 1, 2025, with Vice President JD Vance breaking the tie and Republicans Susan Collins, Rand Paul and Thom Tillis voting no. The House took the Senate’s version 218-214 two days later, with Brian Fitzpatrick and Thomas Massie the only Republican defectors. Trump signed it on July 4. It is Public Law 119-21, marketed as the One Big Beautiful Bill Act.
CBO analyzed how the enacted law would change the resources available to households and published the results Aug. 11. The finding is the load-bearing fact in this entire argument. Resources would decrease for households toward the bottom of the income distribution and increase for households in the middle and toward the top.
The agency sorts households into deciles by size-adjusted income after transfers and taxes. By 2034, CBO projected, the lowest decile would see resources fall by about 3.1% while the highest decile gained about 2.7%.
That is a considerably more precise claim than “tax cuts for the rich,” and it’s harder to argue with, because it doesn’t require anyone to accept a characterization of motive. It’s a projection of where money ends up, produced by the scorekeeper Congress created for exactly that purpose.
Two ledgers, pointed in opposite directions
CBO traced the effect through four channels over 2026 to 2034.
Federal taxes and cash transfers increase household resources by $3.3 trillion on net, driven by the permanent extension of 2017 tax act provisions and by new limits on eligibility for Affordable Care Act insurance subsidies. Federal and state in-kind transfers decrease household resources by $900 billion, primarily because federal spending on Medicaid and SNAP benefits falls. States’ own fiscal responses add back $11 billion. A fourth category CBO allocates as if it were a public good, covering defense, border security and infrastructure, adds $308 billion.
Net those figures and you get a number that sounds fine. Households come out ahead on paper.
But the channels don’t reach the same households. Tax and cash-transfer changes flow disproportionately toward people who pay meaningful amounts of federal tax. In-kind benefit reductions land on people who qualify for means-tested programs. Those are different populations, mostly, and running one up while running the other down is what produces the upward shift CBO describes.
That’s the whole mechanism. It isn’t hidden and it isn’t subtle. Two ledgers, pointed in opposite directions, and the people on each ledger are not the same people.
Medicaid
Saying Trump cut health care is too broad to survive contact with anyone who knows the details.
What the law did was reduce federal Medicaid spending and impose eligibility-related changes that CBO folded directly into its distributional analysis. The largest is a community engagement requirement: adults aged 19 to 64 in the Medicaid expansion population must document 80 hours a month of work, job training, schooling or community service, with states implementing by the end of 2026 and a possible extension to the end of 2028. States must redetermine expansion enrollees’ eligibility every six months instead of annually. The law freezes provider taxes and steps the hold-harmless threshold down from 5.5% in fiscal 2028 to 3.5% in fiscal 2032. Beginning in October 2028, states must charge expansion adults above the poverty line up to $35 per service, with primary care, mental health and substance use treatment exempt.
CBO scored the work-reporting provision alone at $325.6 billion in reduced federal Medicaid outlays through 2034 and projected it would leave 5.3 million more people uninsured in 2034. Across all of the law’s health provisions, the agency put the increase at 10 million uninsured in 2034 relative to its January 2025 baseline, with 7.5 million of that coming from the Medicaid and CHIP changes.
The distributional consequence follows from who Medicaid covers. The program serves lower-income and medically vulnerable people almost by design. Reducing its benefits therefore does something different, in distributional terms, from reducing a tax paid mostly by high earners. One shrinks resources at the bottom. The other expands them at the top.
You can support the underlying policy and still concede that arithmetic. The two positions aren’t in conflict.
The state match
SNAP requires the same care. The law did not abolish the program, and arguing that it did hands the other side an easy correction.
What it did was shift cost. Beginning in fiscal 2027, states pay 75% of SNAP administrative costs instead of 50%. Beginning in fiscal 2028, states with payment error rates above 6% pay between 5% and 15% of actual benefit costs, a share the federal government has carried alone since the program began. States with error rates above 13.3% can delay until 2029 or 2030. CBO expects most of the federal savings from that provision, roughly $35 billion of $41 billion, to come from states writing the checks instead.
Follow the chain. Washington shifts a share of the cost to a state capital. The state legislature, which generally has a balanced-budget requirement, has to close the gap. It raises a tax, cuts a different program, tightens SNAP administration, or leaves the program. CBO modeled all of those responses and estimated the matching requirement alone would end benefits for about 300,000 people in an average month. Separately, the law’s expanded work-reporting rules for able-bodied adults would cut participation by roughly 2.4 million people in an average month over the decade.
Whether the program’s name stays in the federal budget is not the question. Whether fewer resources reach the households that depend on it is the question, and CBO answered it.
Who actually pockets a business tax cut
The claim about corporations needs to be separated from the claim about wealthy households, because the evidence for each is different in kind.
For households, CBO’s distributional analysis does the work directly. For businesses, it doesn’t. The law leaves the corporate rate at 21% and instead makes a set of business provisions permanent: 100% bonus depreciation for qualifying property placed in service after Jan. 19, 2025, immediate deduction of domestic research and experimentation costs, a 20% deduction for qualified business income from pass-through entities, and a new 100% write-off for certain manufacturing structures through 2030. The Joint Committee on Taxation scored permanent bonus depreciation alone at roughly $363 billion over 10 years.
There’s a genuine analytical problem here that people on the left tend to wave past. When a tax break is formally granted to a business, the economic benefit doesn’t stay there. It gets distributed among shareholders, workers and customers in proportions that economists argue about at length and don’t agree on. Shareholding is heavily concentrated, which pushes the benefit upward; the Federal Reserve’s distributional accounts put more than 87% of corporate equities and mutual fund shares in the hands of the wealthiest 10% of households. Wage and price effects push some of it elsewhere. Anyone who tells you the incidence question is settled is selling something.
The disciplined claim is about who captures the benefit, and it requires evidence about that specific question rather than an assumption that a business provision equals a windfall for the wealthy.
The deficit
Two claims get conflated here constantly, and only one of them is true.
Trump did not create the federal deficit. The United States ran enormous structural deficits long before January 2025, driven by Social Security, Medicare, Medicaid, defense, interest on the debt, tax expenditures, prior legislation, demographic change and the ordinary business cycle. Attributing the national debt to any single president is not an argument, it’s a bumper sticker.
The defensible version is narrower. Deficit reduction was a stated fiscal objective of the administration and its congressional allies, and the major legislation they passed moved the number the other way. CBO estimated that PL 119-21 will increase the deficit by $3.4 trillion over 2025 to 2034, the product of a $1.1 trillion decrease in direct spending against a $4.5 trillion decrease in revenues. Add the $718 billion in additional interest the government will pay on the borrowing, and the cumulative effect reaches $4.1 trillion. The law also raised the debt ceiling by $5 trillion, to $41.1 trillion.
He inherited the deficit. He also isn’t absolved of what happened to it afterward.
What the money bought instead
Federal budgeting is a set of competing claims on the same pool of money.
The law directed $150 billion to the Defense Department, including a $25 billion start on the Golden Dome missile defense program and $26.5 billion for shipbuilding. It directed roughly $170 billion, available through September 2029, to immigration and border enforcement, including about $45 billion for detention capacity, close to $47 billion for border wall construction and about $30 billion for Immigration and Customs Enforcement personnel and operations.
Nobody has to accept that Medicaid and food assistance are morally superior claims on the Treasury. The point is that choosing the other list was a choice, made by identifiable people who recorded their votes, and it produced a measurable result that CBO went ahead and measured.
The four strongest objections
Biden caused inflation. Inflation did surge on his watch, and pretending otherwise is a losing position. It was also a global phenomenon following pandemic disruptions, supply shocks and the energy consequences of the war in Ukraine, and U.S. fiscal policy contributed to demand on top of that. How much is genuinely disputed among economists who have no political stake in the answer. The concession to make is that inflation was a real failure of that presidency. The point to hold is that the end-of-term reading was 2.9%, not 9.1%.
Trump inherited a terrible economy. Ask which metric. If terrible means prices relative to 2019, that’s fair. Housing affordability, fair. Interest rates, fair. If it means an unemployment crisis or collapsing payrolls, the December numbers say otherwise, and 4.1% is not a recession reading by any standard definition.
Tax cuts stimulate growth. Maybe. That’s an argument about aggregate economic effects, and it’s a legitimate one; the Tax Policy Center estimates the law raises GDP in 2035 by about 0.5%. CBO’s distributional analysis is an argument about who receives resources. A policy can grow the economy and still move resources upward, and pointing at growth doesn’t answer the distributional question.
Medicaid and SNAP reductions are necessary to control spending. Also a legitimate premise, and it deserves a straight answer rather than a reflex. If deficit reduction is the objective, then the thing to explain is why the same law that reduced low-income assistance also cut revenues by $4.5 trillion, and why the net fiscal effect went the direction it did.
Three questions, in order
What did he inherit? A labor market at 4.1% unemployment, inflation down to 2.9% but still above target, a heavily funded military carrying two decades of accumulated readiness problems, high prices, high rates and a structural deficit.
What did he change? Tax law and federal spending, including Medicaid and SNAP, through a reconciliation bill that cleared the Senate by one vote and the House by four.
Who gained and who lost? CBO’s answer is on the record. The bottom decile loses about 3.1% of its resources by 2034. The top decile gains about 2.7%.
That’s the argument. Not that a perfect economy was handed over and destroyed. That relatively favorable conditions were inherited, and federal policy was then used to shift resources in a direction the government’s own nonpartisan scorekeeper described in advance, in writing, before the vote.
Somewhere in the next two years a state legislature is going to open its session with a SNAP matching bill and a revenue shortfall on the same agenda. Whatever it decides, the households on the receiving end will not have been in the room.