A total you can never pay is an abstraction. A payment that comes due every year, ahead of almost everything else in the budget, is a constraint. This page tracks the second one.
Nobody expects the federal government to retire $39.8 trillion. Sovereign debt is rolled, not paid off, and that is precisely why the total on its own has never been a useful investment signal. It has been climbing and alarming for four decades, through expansions and contractions alike. Anyone who repositioned a retirement portfolio around it in 1990, or 2000, or 2012 gave up far more than they protected.
Interest is a different kind of number, because interest is a payment. It clears before discretionary spending. It does not wait on an appropriations fight or respond to an election. For fiscal 2026 the Congressional Budget Office puts net interest at roughly $1.0 trillion — more than the federal government spends on national defense, more than Medicaid, more than veterans' benefits and services.
Two ratios say it better than the raw dollars. Interest now absorbs about 18.6 percent of all federal revenue, above the previous record set in 1991, and CBO's projection has that reaching roughly a quarter of revenue by 2036. As a share of the economy, interest costs have run from 1.6 percent of GDP in 2021 to a record 3.2 percent in 2025 and higher since.
The reason this belongs on an adviser's website is narrow and specific, and it is worth stating plainly before anything else on this page: a rising interest bill is not a prediction of a crash. It is one of the inputs that sets the plausible range for the cost of money. And the cost of money is what prices every long-duration asset a retiree owns — the bond ladder, the utility stock whose allowed return moves with rates, the growth equity whose value sits mostly in cash flows a decade out.
Most of the debt outstanding today was issued when money was nearly free. Across all its interest-bearing securities, the weighted average rate the Treasury actually pays was roughly 1.5 percent in 2021. As of 30 June 2026 it was 3.41 percent, and it is still drifting upward — because the cheap paper from 2020 and 2021 keeps maturing and getting refinanced at whatever the market charges now.
This is the part that surprises people. Even in a hypothetical year where Congress ran no deficit at all and added nothing to the pile, the interest line would keep climbing for a while, because the average coupon on the existing stock is still catching up to the market. Rolling debt is not free when the rate you roll into is more than double the rate you rolled out of.
It runs the other way too, and that is the honest half of the mechanic. If rates fall materially and stay down, the same arithmetic reverses and the interest line decelerates. Nothing here is a one-way ratchet. What has changed is that the outcome now depends on the path of rates to a degree it did not when the average coupon sat at 1.5 percent.
Interest has passed national defense. It has passed Medicaid. It is now running level with Medicare, and the Congressional Budget Office projects it moves clearly past Medicare within the next couple of years — leaving only Social Security ahead of it. Set against the other major spending categories for fiscal 2026, it looks like this.
A dollar spent on interest buys nothing. It is not a road, a carrier, a hospital bed or a benefit check. It is the cost of decisions already made, and it has first claim. That is what makes it a structural item rather than a political one: every other line on the page can be argued about, and this one arrives on schedule regardless.
I am not going to put a date on this, and I am not going to tell you a rising interest bill is a reason to sell anything. What I will say is that it belongs in how a retiree sizes a cash reserve — because a rising cost of money makes the early years of a bad sequence worse, and the early years are the ones that do permanent damage.
A debt clock is the single most-used alarm device in financial marketing, and as a timing signal it has been wrong for forty consecutive years. These are the objections that deserve to be made at full strength, not softened.
Every decade, someone has pointed at a rising debt figure and called a top. A retiree who moved to cash on that basis in 1995, or 2010, or 2019 would be measurably poorer today, and would have spent thirty years being right in principle and wrong in outcome. Being able to describe a pressure is not the same as being able to time it.
Unlike a household, or Greece, the United States cannot be forced into an involuntary default on dollar obligations. The binding constraint is inflation and the political tolerance for it, not solvency in the ordinary sense. That makes the family-budget framing which usually accompanies a debt clock genuinely misleading, and I would rather say so than lean on it.
Japan has run debt above 200 percent of GDP for years without the crisis repeatedly forecast for it. Work out of Penn Wharton puts a US outer bound somewhere around 200 percent, and capable economists dispute that figure in both directions. Anyone who names the threshold precisely is guessing, and that includes anyone naming it on this website.
The rollover argument above is symmetric. A sustained decline in rates would decelerate the interest line exactly the way the increase accelerated it. The $1.0 trillion figure is CBO's projection, not mine, and CBO's rate assumptions have missed in both directions before.
Debt and interest are usually stated against GDP. A genuine, sustained productivity increase — the case being made for the current capital investment cycle — would improve these ratios without a single policy decision. That has happened before, most obviously in the decades after 1945, when a debt load larger relative to the economy than today's was outgrown rather than repaid.
Taken together, these do not dissolve the argument. They locate it. A rising interest bill is a statement about the structure a portfolio has to survive, not a statement about what happens next quarter. Anyone using it as a market call — including anyone using this page that way — is using it wrong.
Macro sets structure. It does not set timing. If a rising interest bill changes anything about a retirement plan, it changes these five things — none of which require a view on what the market does next.
Two to three years of planned withdrawals held in something that never has to be sold at a loss. This is a buffer, not a forecast — it costs a little in expected return and buys the ability to not sell into a decline.
If the path of rates is genuinely less predictable than it was, a long-duration bond position is a directional bet whether or not it was intended as one. Worth knowing which bet is on the table.
Inflation is the channel through which fiscal pressure most plausibly reaches a household. A pension that is not indexed is a specific exposure to that channel. Social Security, which is indexed, is not.
For a utility household this is the one that matters. The allowed return on a regulated utility moves with the cost of money. Wage or pension, retiree medical, and a concentrated equity position can all rest on the same rate environment and the same balance sheet at once.
The real value of thinking about any of this in a calm year is that it lets you decide beforehand what you would do, at what level, and commit to it — instead of deciding in the middle of a decline, which is when people make the decisions they later regret.
If you are within a few years of a pension election, or holding a concentrated position in the company you worked for, the useful conversation is not about the national debt. It is about how much of your own plan depends on the cost of money staying where it is.
A fee-only, state-registered investment adviser working with utility-industry families on concentration, pension elections and sequence-of-returns risk.
Every figure on this page comes from one of the following. Where a number will date, the date it was read is stated beside it in the text.
Bailey Financial Services, Inc. is a state-registered investment adviser. Registration does not imply any level of skill or training, and nothing on this page is an offer or solicitation in any jurisdiction where the firm is not registered or exempt.
This page is educational and general in nature. It is not investment, tax or legal advice, and it is not individualized to any person's circumstances. Nothing here should be read as a forecast of markets, interest rates or the economy, or as a recommendation to buy, sell or hold any security. No date is placed on any of it deliberately.
About the counters: the total debt figure reads from the U.S. Treasury's Debt to the Penny dataset, which posts once each business day. The interest readouts are not a live measurement. They are a straight-line estimate derived from the Congressional Budget Office's projection of approximately $1.0 trillion in net interest for fiscal 2026, divided evenly across the year. Actual interest outlays are lumpy, are reported monthly rather than continuously, and will differ from the figures shown. The counters are illustrative of a pace, not a statement of an amount owed at any instant.
Third-party data, publications and organisations referenced here are cited as sources only. Their inclusion is not an endorsement of Bailey Financial Services by them, or of them by the firm. Bailey Financial Services is not affiliated with, endorsed by, or sponsored by Southern Company, Georgia Power, or any utility, and any reference to utility employment is descriptive of the clients the firm serves.
Historical patterns and ratios discussed here are descriptions of the past. They are not predictions, and past performance does not indicate future results. All figures were current as of the dates noted and are not updated continuously except where explicitly identified as a live feed.