The Fiscal Story We Have Not Finished Telling
By Jonah Ritter · September 16, 2026
A 10-year Treasury yield near 5% gets my attention. The last time we saw a rate around this level was before the financial crisis. That does not mean another crisis is imminent. But it does raise a fair question: Are investors demanding a higher return because they expect stronger growth, because inflation may persist, or because lending to a heavily indebted government for ten years now requires a bigger premium?
The answer matters. A 10-year yield reflects expectations about future short-term rates, inflation, and compensation for holding a long-term bond. It is not a direct reading of today's inflation rate. Nor does every dollar of existing federal debt immediately refinance at the new yield. But as maturing debt is replaced and new deficits are financed, higher rates steadily raise the government's interest bill.
That creates the feedback loop I worry about: more interest expense adds to the deficit; the deficit adds to the debt; and a larger debt can push interest expense higher still. The danger grows if investors begin to doubt that economic growth and federal revenue can keep pace.
The deficit Trump walked into
We should start with the calendar. On January 17, 2025—before President Trump took office—the Congressional Budget Office projected a $1.9 trillion deficit for fiscal 2025, a fiscal year that was already more than three months underway. It also projected federal debt held by the public rising from about 100% of GDP in 2025 to 118% in 2035 under the laws then in place. Those projections show that the debt and much of the 2025 deficit were inherited. They do not mean every subsequent outcome belongs to a previous administration. CBO's January 2025 outlook.
By February 2026, CBO still projected an approximately $1.9 trillion deficit for fiscal 2026 and debt reaching 120% of GDP in 2036. Its update also shows why attribution has to be honest: CBO estimated that the 2025 reconciliation law and immigration-related administrative actions increased projected cumulative deficits, while higher tariffs reduced them. A president inherits a budget trajectory and then changes it, for better or worse, through policy and economic results. CBO's February 2026 outlook.
Why cuts can take time to appear
An announcement is not the same as a lower Treasury payment. Some employees who accepted the administration's deferred resignation offer continued receiving pay and benefits through September 30, 2025. In those cases, the payroll benefit would begin later. Other personnel reductions may involve severance, accrued leave, continuing benefits, or work shifted to contractors. The relevant figure is the net reduction in actual spending, not the number of departures announced. Office of Personnel Management's deferred resignation FAQ.
Fraud and waste work the same way. Identifying an improper payment does not instantly put cash back in the Treasury. Savings appear when future payments are stopped; recoveries appear when money is actually collected. Some discoveries may prevent large future losses even if the deficit in the discovery year barely changes. And an estimated improper payment is not automatically proof of fraud or a sum that can all be recovered.
There is also a scale problem. Federal civilian payroll is one component of a budget dominated by Social Security, health programs, defense, and interest. Genuine administrative savings matter, but they must be large and sustained to move a deficit measured in trillions.
What is happening beneath the headline?
One useful measure is the primary deficit: the deficit with net interest costs removed. It tells us whether current revenue covers current noninterest spending. It is a clearer way to spot operating improvements, although it does not make interest disappear from the government's actual bills.
CBO projects a primary deficit of 2.6% of GDP for fiscal 2026. Its August monthly review shows that, through the first eleven months of the fiscal year, net interest outlays were $111 billion higher than in the comparable 2025 period. After adjusting for shifts in payment dates, the primary deficit was modestly lower than a year earlier. That offers some evidence of improvement beneath the headline, but not enough yet to establish a lasting turnaround.
There is a legitimate case for patience: inherited commitments, payroll arrangements, contracts, and enforcement changes take time to show up in cash spending. But 2028 is a testable hypothesis, not a guaranteed payoff date. We should expect to see measurable improvement in the primary deficit and debt relative to the economy as those changes mature.
What the deficit builds
The deficit tells us how much the government must borrow. It does not tell us whether the money paid for ongoing expenses or helped build something the economy can use for years. Congress provided $12.5 billion to begin modernizing the air traffic control system, replacing aging equipment and facilities. In September 2026, the Transportation Department announced $1.1 billion in airport grants for projects including runways, terminals, a control tower, and equipment that helps controllers track vehicles on the ground. Construction also began this year on the replacement Blatnik Bridge, a freight and commuter link between Minnesota and Wisconsin, using a federal grant awarded in 2024. FAA; Transportation Department airport grants; Transportation Department bridge announcement.
These projects still cost money, and their value depends on finishing them efficiently. But safer flights, more reliable bridges, and less congested airports can support commerce long after the construction bills are paid. When we judge federal spending, we should ask not only “How much did it cost?” but also “What did the country gain?”
Signs the economy may have more room to grow
There is also good news that deserves attention. In the second quarter of 2026, real GDP grew at a 1.5% annual rate, while real final sales to private domestic purchasers grew 4.2%. That second measure captures household spending and private fixed investment and suggests firmer underlying private demand than the headline GDP number alone. It is one quarter, not a settled growth trend. Bureau of Economic Analysis.
Capital is still coming into American businesses. Foreign investors spent $232.2 billion acquiring, establishing, or expanding U.S. businesses in 2025, up 49.5% from 2024; manufacturing accounted for 52.5% of the total. Most of that money purchased existing businesses, so it would be misleading to call the entire amount new factories. The longer-term opportunity is for ownership, new plant construction, equipment spending, and supplier investment to turn into more output here. Bureau of Economic Analysis.
The 2025 investment figures do not capture the scale of what has since been announced. President Trump has spoken of trillions of dollars in planned U.S. investment, and the "White House lists commitments" from American companies and foreign governments across manufacturing, technology, energy, and infrastructure. In 2026, the United States and Japan selected the first projects under Japan’s $550 billion investment commitment, showing how a broad pledge can begin to take shape in specific projects. This is a substantial potential source of future growth. The announced totals cover different years and types of commitments, however, including planned spending and purchases; they are not a measure of money already invested or factories already producing. The test is how much becomes financed construction, equipment, jobs, and output in the United States. U.S. Department of Commerce."
New factories and AI infrastructure can expand capacity, and AI tools could raise what each worker produces. Those gains would help businesses earn more, pay higher wages, and generate tax revenue without requiring a higher tax rate. But productivity gains are earned through deployment and actual output, not through announcements of future projects. Manufacturing construction has been substantial, though recent construction data also show a slowdown from its earlier pace. Federal Reserve monetary policy report; Census construction spending release.
Allied defense spending adds another potential source of demand for American manufacturing. NATO reports that European allies and Canada increased core defense expenditure by nearly 20% in 2025 compared with 2024. Allies have also committed to invest 5% of GDP in defense and related security needs by 2035, including at least 3.5% for core defense. That creates opportunities for U.S. producers and suppliers of air defense systems, aircraft, electronics, drones, and emerging autonomous technologies. Successful exports could support factory expansion, skilled jobs, research, and earnings here at home. The spending targets do not guarantee U.S. orders: allies may buy from their own industries, and contracts take time to become production and revenue. NATO's 2026 spending update; The Hague Summit Declaration.
Public safety belongs in the economic picture
Crime takes a toll that a GDP table cannot fully show. Violence destroys lives and property; addiction tears through families and workplaces; fear changes where people shop, invest, hire, and raise children. Safer communities can make it easier for businesses to operate and for people to participate in the economy. That benefit matters even when it cannot be converted into a neat deficit estimate.
The border numbers changed quickly after the administration took office. DHS reported that encounters with gotaways—people detected crossing who evaded apprehension—were down about 95% during Trump's first 100 days. In June 2026, DHS reported 13 consecutive months with no Border Patrol releases at the border. That describes a specific release practice for people apprehended by Border Patrol; it does not mean no one entered the country unlawfully or that every type of admission ended. The abrupt change is an important result of the administration's border policy, though a claim about its precise effect on nationwide crime would require separate evidence. DHS first-100-days report; DHS June 2026 release.
The broader public safety numbers are striking. The FBI estimates that violent crime fell 9.3% in 2025, the largest one-year decline in its national violent-crime rate estimates since they began in 1936. Murder and nonnegligent manslaughter fell 18.1%, robbery 18.5%, aggravated assault 7.2%, and rape 7.6%. Property crime fell 12.4%, including a 22.7% drop in motor vehicle theft. The estimated murder rate of 4.1 per 100,000 tied the rates recorded in 1955 and 1956. These are national estimates based on crimes reported to participating agencies. FBI 2025 crime report; FBI account of the decline.
Drug deaths offer another measure of human and economic relief. The CDC's provisional estimates show opioid-involved overdose deaths fell about 19%, from 55,296 in 2024 to 44,564 in 2025. DEA testing found that the share of fentanyl pills it analyzed containing a potentially lethal dose fell from 76% in fiscal 2023 to 29% in fiscal 2025. That is a comparison of tested pills, not proof that all fentanyl sold in America became equally less dangerous. The DEA also reports major seizures, although seizure totals alone cannot tell us how much drug supply ultimately reached users. CDC provisional 2025 data; DEA fentanyl testing and seizures.
We should give credit for the sharp border shift and take the 2025 crime decline seriously. We should also get the timeline right: overdose deaths were already falling before January 2025, and a nationwide crime total does not by itself identify which federal, state, or local policy caused the decline. That is precisely why critical thinking matters. When we look only at the deficit, we miss gains in safety and human welfare. When we attribute every favorable statistic to one election, we miss the evidence that would tell us which improvements can last. CDC on the earlier overdose decline.
The growth case—and its test
A strong economy can make a large debt easier to carry because the denominator—GDP—grows, incomes and tax receipts rise, and the government has more capacity to service what it owes. That may be one reason some investors are willing to give the fiscal story more time. Still, growth alone cannot be assumed to outrun persistent primary deficits and rising interest costs. Strong demand can also keep inflation and long-term rates elevated; the second-quarter PCE price index rose at a 5.3% annualized rate, a reminder that the inflation question has not vanished. Bureau of Economic Analysis.
The 10-year yield may be telling us several things at once: America has real prospects for investment and growth, while lenders want greater compensation for inflation uncertainty and a large supply of government debt. We cannot know the precise mix from the yield alone.
I am willing to give delayed savings and new investment time to work. But the evidence I want to see is concrete: sustained productivity and private investment, a falling primary deficit, interest costs stabilizing relative to federal revenue, and debt no longer growing faster than the economy. If those measures improve, a 5% Treasury yield will look more like the price of a growing economy. If they do not, it will look more like a warning we should have taken seriously.
