The U.S. national debt stood at roughly $38 trillion in late 2026, measured as total public debt outstanding — the gross figure that includes money the government owes itself. That's the number the Treasury's daily Debt to the Penny tracker reports. The more economically meaningful figure, debt held by the public, runs several trillion dollars lower, near $30 trillion. The gap between the two explains most of the confusion you'll see in headlines, and it's the first thing to pin down before quoting any figure at all.

What the $38 Trillion Gross Figure Actually Counts

Gross federal debt is the whole balance sheet item: every Treasury security outstanding, whether a pension fund, a foreign central bank, or the Social Security trust fund holds it. The Treasury's Bureau of the Fiscal Service publishes this daily through Debt to the Penny, and it moves by tens of billions on any given business day. That's not noise — it's issuance, redemption, and interest accrual all hitting the ledger at once.

Here's the part most explainers skip. About $7 trillion of the gross figure is intragovernmental debt, money the government owes to its own trust funds, chiefly Social Security and Medicare. That debt is real in a legal sense and completely real for the trust funds' accounting. It is not money owed to the public, and it doesn't trade in markets, so it doesn't compete with private borrowing for capital.

Run the arithmetic yourself: $38 trillion gross minus roughly $7.5 trillion intragovernmental puts you near $30.5 trillion held by the public. Two honest people citing "the national debt" can differ by $7 trillion and both be right. The conflict isn't a dispute about facts; it's a dispute about which boundary someone drew.

One practical test when you see a figure in the wild: ask whether it includes trust fund holdings. If the article doesn't say, assume gross, because that's the larger, scarier number and it gets quoted more. The Treasury's own Monthly Statement of the Public Debt breaks the total into its components if you want to verify a specific claim.

Debt Held by the Public: The Number Economists Prefer

Debt held by the public is the figure that matters for credit markets, interest-rate pressure, and international comparisons. It's what foreign governments own when they hold U.S. Treasuries — Japan and China have long been the two largest foreign holders, each in the hundreds of billions to low trillions depending on the year. It's also what the Congressional Budget Office uses when it projects deficits and their economic effects.

The distinction has real market consequences. Intragovernmental debt doesn't need to be refinanced in public auctions, so it doesn't affect Treasury auction demand or yields. Public debt does. When the Treasury sells securities to cover a deficit, those sales compete for the same capital as corporate bonds and mortgages, which is why economists watch the public figure when discussing crowding out.

If you're comparing the U.S. to another country — say, asking whether U.S. debt is "worse" than Japan's — you must use public debt as a share of GDP for both. Japan's gross debt is famously above 250% of GDP, but a large share of it is held domestically by the Bank of Japan and government institutions. Mixing gross for one country and public for another produces nonsense.

As a rough benchmark, U.S. debt held by the public in late 2026 is somewhere in the neighborhood of 100% of GDP, depending on the GDP vintage you use. The exact ratio shifts quarterly with GDP revisions, so treat any precise percentage as a snapshot with a date attached, and check the CBO's latest baseline for the current figure.

Why You'll See Four Different Numbers in Circulation

Search the phrase and you'll encounter at least four distinct figures, each with a legitimate source behind it. Gross debt, public debt, debt subject to limit, and net debt all describe slightly different things. Knowing which one a headline used is the whole skill.

Measure Approximate late-2026 level What it includes Primary source
Gross federal debt ~$38 trillion All Treasury securities outstanding Treasury Debt to the Penny
Debt held by the public ~$30.5 trillion Securities held outside government Treasury Monthly Statement
Intragovernmental debt ~$7.5 trillion Trust fund holdings Treasury Monthly Statement
Debt subject to the statutory limit ~$38 trillion Nearly all gross debt Treasury / Congress

Notice that debt subject to the limit and gross debt are close but not identical — a small sliver of gross debt sits outside the statutory ceiling. That's why a "debt limit" fight can be resolved while the gross figure keeps climbing the next morning. The ceiling applies to the measure Congress set, not to every obligation.

Which should you quote? Match the audience. For a general "how big is it" answer, gross is conventional. For policy or market analysis, public debt. For a debt-ceiling story, debt subject to limit. Stating which one you mean, in the sentence where you use it, eliminates most of the ambiguity.

The Pace of Growth: What a Decade of Increases Looks Like

The number is not static, and the rate of change matters as much as the level. Gross debt crossed $20 trillion around 2017, $30 trillion in early 2022, and has been climbing toward $40 trillion through the mid-2020s. That's roughly a $10 trillion increase every few years, faster than nominal GDP growth in several of those stretches.

CBO baseline projections put annual deficits in the $1.5 trillion to $2 trillion range across the coming decade under current law, which implies continuous upward pressure on the debt. Interest costs are the fastest-growing piece: as the debt rolls over at higher rates, net interest expense has climbed past defense spending in the budget, a crossover that changes the arithmetic of every future fiscal debate.

A worked example of why that compounds. Suppose $30 trillion of public debt carries an average rate of 3.5%. That's roughly $1.05 trillion in annual interest — computed as 0.035 × 30,000, a figure close to what the Treasury reports for net interest outlays in recent years. Raise the average rate to 4%, and the same debt costs $1.2 trillion. The debt didn't change; the rate did; the budget did.

Edge case worth flagging: the gross figure can spike on a single day for technical reasons, like a large trust fund reinvestment or Treasury's cash management around quarterly refunding. A one-day jump doesn't mean the government borrowed that much in new money. Use monthly or quarterly averages for trend statements.

Foreign Holders, Trust Funds, and Who Actually Gets Paid

Follow the money and the "who do we owe" question gets more interesting than the politics suggest. Foreign investors hold a substantial share of public debt, but the largest single holder is usually the Federal Reserve, which accumulated Treasuries through quantitative easing. Fed holdings affect the market differently from foreign holdings — the Fed can let securities mature without reinvesting, which tightens conditions without any foreign policy dimension.

Domestic investors — mutual funds, banks, pension funds, state and local governments, individuals through savings bonds and money market funds — actually hold the majority of publicly held debt when you aggregate them. The "we owe it to China" framing captures a real slice but overstates the foreign share, which has generally run somewhere between a quarter and a third of public debt in recent years.

Then there's the trust fund layer. When Social Security runs a surplus, it buys Treasury securities, and the government counts those as intragovernmental debt. When the program runs a deficit, as it has been, it redeems those securities and the Treasury has to borrow from the public to cover the redemption. That's the moment intragovernmental debt stops being invisible — it converts into public borrowing.

If you're trying to reason about whether the U.S. can "afford" its debt, the holder mix matters. Debt owed to domestic residents in domestic currency is a different constraint than debt owed abroad in foreign currency. The U.S. borrows in dollars it issues, which is why default risk is discussed as a political failure mode, not a solvency one, by most mainstream economists.

What Each American's Share Actually Means

Divide gross debt by population and you get roughly $110,000 per person — $38 trillion across about 340 million residents, computed as 38,000 divided by 0.34. Divide public debt instead and you land nearer $90,000. Both numbers get thrown around as shock statistics, and both are arithmetically honest given their inputs.

Here's where the per-capita framing breaks. Every American doesn't owe that amount, because much of the debt is held by Americans themselves through funds, pensions, and savings. A household holding a Treasury bond in a retirement account is on both sides of the ledger. The per-capita figure answers "how much debt exists per resident," not "how much will each person write a check for."

What you can use the per-capita figure for is scale comparison. Federal spending on any given program, measured in the tens of billions, looks small against a debt measured in tens of trillions. A $50 billion program is roughly 0.13% of gross debt. That ratio is the honest way to express "this is small relative to the debt" without hand-waving.

If you want a genuinely useful per-household metric, use debt held by the public divided by households — roughly 131 million households in the U.S., which puts public debt around $230,000 per household as of late 2026. That's the burden the tax base ultimately services, and it's a fairer denominator than raw population for fiscal discussions.

The Debt Ceiling: Why the Statutory Limit Isn't the Debt

The statutory debt limit is a congressional cap on how much the Treasury can borrow, and it has been raised or suspended dozens of times since it was created in 1917. It is not a spending authorization — Congress has already appropriated the money by the time the ceiling binds. That mismatch is the source of the recurring standoffs.

When the limit is reached, Treasury runs down its cash balance and uses "extraordinary measures" — suspending reinvestment in certain trust funds, for example — to keep paying bills. These measures buy weeks or months, not years. If Congress doesn't act, the government can't legally issue new debt to cover obligations already incurred, which is the default scenario analysts debate.

The number to know is the debt subject to limit, which tracks gross debt closely. Because it moves daily, any figure you read ages immediately. Treasury publishes it in the Daily Treasury Statement, which is the authoritative source; check there rather than trusting a headline number.

A misconception worth correcting: raising the ceiling doesn't authorize new spending. It permits payment of obligations Congress already created. The spending decisions happened in appropriations bills and tax laws, sometimes years earlier. The ceiling vote is about whether to honor those commitments, which is why most economists treat brinkmanship over it as a self-inflicted risk.

How to Check the Current Number Yourself in Under Two Minutes

Don't rely on any article, including this one, for a live figure — the number changes every business day. Go to the source and read it directly. The process is short.

1: Open the Treasury's Fiscal Data site and locate the "Debt to the Penny" dataset. It updates each business day and shows the total, the public component, and the intragovernmental component. 2: Read the "Total Public Debt Outstanding" line. That's the gross figure. The two sub-lines underneath give you public and intragovernmental debt without any subtraction on your part. 3: For the debt ceiling measure, open the Daily Treasury Statement and find the "Total Public Debt Subject to Limit" line, which is usually near the bottom of the summary table. 4: For context on trajectory, open the CBO's most recent Budget and Economic Outlook and check its baseline projections for deficits and debt held by the public over the next ten years.

If the site won't load or the dataset shows a stale date, check whether the last update fell on a holiday or weekend — Treasury doesn't post on non-business days, so the most recent figure may be two or three days old. That's normal, not a sign the data is broken.

One failure case to know: during debt-limit episodes, Treasury sometimes publishes a note explaining that it's using extraordinary measures. The headline number may appear to plateau or move in unusual increments. If you see those conditions, the number is still accurate as a balance, but it's not a comfortable snapshot of normal borrowing.

Where the Debt Goes From Here: Baseline Projections and Their Limits

CBO's long-term outlook has public debt rising as a share of GDP over the coming decades under current law, driven mainly by an aging population, health care costs, and interest on the debt itself. That's a projection, not a forecast, and it depends heavily on assumptions about rates, growth, and legislation that hasn't been passed. Every one of those assumptions is contestable, which is why serious analysts present scenarios rather than point estimates.

The practical implication for anyone quoting a number: attach a date and a definition. "The debt was $38 trillion in November 2026, measured as gross federal debt" survives scrutiny. "The debt is $38 trillion" invites an argument about whether you meant gross or public, and about what month you were reading. Precision costs a clause and saves the whole conversation.

Watch for the interest-cost crossover in any future update. When net interest exceeds major program spending, the debt becomes self-reinforcing — borrowing to pay interest raises the debt, which raises interest. That threshold has been crossed in recent budgets, and it's the single most important number for understanding why the trajectory worries economists more than the level does.

If you're making a decision that depends on the debt — a portfolio allocation, a policy argument, a paper — name the measure, the date, and the source in one sentence, then move on. The number will be different next quarter. The framework for reading it won't be.

FAQ

Is the national debt the same as the deficit?

No — the deficit is the annual gap between spending and revenue, and the debt is the accumulated total of past deficits minus surpluses. In late 2026 the annual deficit is running above $1.5 trillion, which adds to the roughly $38 trillion gross debt. Confusing the two produces errors of about 20x in either direction.

Does the U.S. actually owe $38 trillion to foreign countries?

No. Foreign holders own somewhere between a quarter and a third of publicly held debt, not the gross total, and the largest single holder is often the Federal Reserve. Roughly $7.5 trillion of the gross figure is owed to U.S. trust funds, meaning the government owes it to itself in an accounting sense.

What happens if the debt ceiling isn't raised?

Treasury runs out of legal borrowing room and must rely on cash balances, which would force delayed payments on some obligations. Most analysts treat this as a default scenario with unpredictable market consequences, not a routine event. The ceiling has been raised or suspended dozens of times since 1917, and each standoff has been resolved before the deadline.

Which debt figure should I cite in a paper or report?

Use debt held by the public if you're discussing economic effects, interest rates, or comparisons with other countries. Use gross debt if the context is the total balance sheet or the debt ceiling. State the measure and the date in the same sentence; that's the difference between a citable figure and an arguable one.

How fast is the debt growing compared to the economy?

Gross debt has been rising by roughly $2 trillion to $3 trillion per year recently, while nominal GDP growth has run somewhat slower in some years. When debt grows faster than GDP, the debt-to-GDP ratio rises even without new policy. Check the CBO's latest baseline for the current-year comparison rather than relying on any older figure.

Does the debt ever get paid off?

The U.S. has run surpluses only a handful of times in modern history, most recently from 1998 to 2001, and the debt fell modestly in those years. Paying off the full gross debt is not a realistic policy goal; stabilizing debt as a share of GDP is the target most economists discuss. That's a different and more achievable benchmark.

Is the $110,000 per-person figure accurate?

It's arithmetically correct for gross debt divided by population, but misleading as a burden measure because much of the debt is held by Americans themselves. For a fairer denominator, divide public debt by households, which yields roughly $230,000 per household as of late 2026. Neither number tells you what any individual will pay.

Why do different news outlets report different debt totals?

They're using different measures, different dates, or both. Gross debt and public debt differ by around $7.5 trillion, and the figure changes daily. An outlet quoting an older snapshot can be off by hundreds of billions without any error. Check the definition and the date before comparing two reports.

This article is general educational information about federal fiscal statistics, not a financial, investment, or legal recommendation. Debt figures change daily and projections depend on assumptions that may not hold; confirm current numbers with the U.S. Treasury's Fiscal Data site and the Congressional Budget Office before relying on them. For decisions with financial consequences, consult a qualified financial or tax professional in your jurisdiction.