Two figures. Fifty years. One republic that cannot bring itself to decide.

*   *   *

Two figures.

The first is 31.7 percent.

The second is 99 percent.

***

The first is the share of American household wealth held by the top one percent of households, as of the third quarter of 2025. The second is the share of American GDP represented by federal debt held by the public, as of the close of fiscal year 2025.

• Both are at the highest levels in living memory.
• Both have been climbing for fifty years.
• Both belong to the same political community.
• The public debate sees one or the other. Almost no one looks at them together.

This essay does. It treats each pattern alone. Then together. Then it gives a verdict — in three parts, because the question has three answers, and the citizens of a republic are owed all of them.

What follows is the work.

I. The Wealth Gap, Considered Alone

The data has thickened since my trilogy Capital and the Republic was published.

Here is what the figures show now.

The Federal Reserve’s Distributional Financial Accounts — the cleanest series the country produces on this question — placed the top one percent’s share of household wealth at 31.7 percent in the third quarter of 2025. That is the highest share since the series began in 1989. It represents approximately fifty-five trillion dollars in the hands of the country’s wealthiest 1.3 million households.

• The wealth held by the bottom ninety percent — roughly one hundred and twenty million households — adds up to about the same.
• The share held by the bottom fifty percent — roughly one hundred and sixty-five million Americans — is 2.5 percent of national wealth.

The American concentration of 2026 is within striking distance of where it stood in 1916, before the income tax, before the antitrust statutes worked their effect, before the New Deal built the architecture that brought the figure back into range over the next four decades.

The productivity-pay gap — the working life of the citizen, expressed in numbers — tells the same story.

From 1979 to 2025, net productivity grew approximately ninety percent. Typical worker compensation grew approximately thirty-three. If pay had tracked productivity over that span, the median worker would be making more than sixteen dollars more per hour today. The figures are the Economic Policy Institute’s, drawn from Bureau of Labor Statistics productivity and compensation series. The pattern is everyone’s.

The labor share of national income — the portion of the country’s earnings that goes to wages and salaries rather than to profits and rents — has fallen to the lowest level in the seventy-five-year postwar series.

The surplus is going somewhere. It is not going to the wages of the eighty percent of the workforce who are production and non-supervisory.

There was a time when the connection ran the other way. From 1948 to 1973, productivity and typical worker compensation grew in close parallel. Productivity rose about ninety-seven percent. Worker compensation rose about ninety-one. The economists of the period treated the linkage as a structural feature of the American economy — an article of faith in the postwar consensus that the gains of broad-based growth would be broadly shared.

After 1973, the linkage broke.

The break has now held for fifty-two years.

Long enough that the working life of an entire generation has passed under its terms.

Now: this is where the public debate goes wrong.

The harm from this pattern is not the headline. It is not the yacht in the harbor. It is not the billionaire on the magazine cover. Those are symptoms. The harm operates beneath them, in four places at once.

• Political representation. The political-economy literature of the last two decades has converged on a finding. In any contested policy domain in the United States, the preferences of the top decile of income earners predict legislative outcomes. The preferences of the median voter do not, except where they happen to coincide. The American republic, on this evidence, has become unequal in a structural sense — not in its outcomes alone, but in its voice.
• Democratic erosion. A 2025 cross-national study by Eli Rau and Susan Stokes of the University of Chicago, published in the Proceedings of the National Academy of Sciences, found that income inequality is among the strongest predictors of where and when democracies erode. The mechanism is polarization. Concentrated wealth concentrates grievance. Grievance splits the political community along economic lines that map onto cultural and partisan ones. The result is an electorate prepared to entertain institutional rupture. The 1930s contained that risk through legislation. The 2020s have not yet matched the legislative response to the data.
• Macroeconomic fragility. In the second quarter of 2025, the top ten percent of income earners accounted for nearly half of all consumer spending in the United States. An economy whose marginal demand is supplied by the top decile is, by construction, more exposed to asset-market reversals than one whose demand is broadly distributed. The K-shape is not a metaphor. It is a balance-sheet condition.

A republic that produces productivity gains its workers do not share will, over time, lose the moral standing on which compliance with the rules of the market rests. That loss is slow. It is also the deepest.

The case for treating the inequality pattern as less urgent than it appears: the United States in 2026 has 4.3 percent unemployment, ninety thousand dollars in GDP per capita, and an asset market that is the envy of the world. The bread lines of 1932 are not in the streets. The patterns are real. They are not, today, producing acute crisis.

The counter-case: the political community has had fifty years to address them, and used the time to make the figures worse.

Time is not a resource it has demonstrated the capacity to use.

II. The National Debt, Considered Alone

Now turn to the debt.

Debt held by the public — the cleanest measure for cross-country and historical comparison — reached approximately ninety-nine percent of GDP at the close of fiscal year 2025. By early 2026, it crossed one hundred percent. Total federal debt, the broader measure that includes intragovernmental obligations, stood at roughly one hundred twenty-three percent.

The previous record, set in 1946 after the costliest war in human history, was one hundred and six percent.

The Congressional Budget Office projects that the 1946 record will be broken in fiscal year 2030. By 2036, debt held by the public will reach one hundred twenty percent. By 2056, on present projections, one hundred seventy-five.

That is the stock figure.

The flow figure is the one to watch.

The federal government paid net interest of $970 billion in fiscal year 2025, equal to 3.2 percent of GDP. That eclipses the previous interest-cost peak set in 1991. CBO projects net interest will reach $1 trillion in 2026 and $2.1 trillion by 2036. Over the decade, the total interest bill comes to $16.2 trillion.

The crossover points are worth setting down each in its own line.

• In fiscal year 2024, the federal government spent more on net interest than on national defense — the first such crossover in modern American history.
• In fiscal year 2028, on current projections, net interest will exceed Medicare.
• In fiscal year 2038, net interest will exceed defense and all non-defense discretionary spending combined.
• By 2048, net interest will be the single largest line in the federal budget. The federal government of 2048 will, on present projections, spend more to service the past than to invest in the future.

Beneath this trajectory sits the figure CBO will not headline. The Committee for a Responsible Federal Budget, working from the February 2026 CBO baseline, calculates that the average interest rate paid on the national debt will exceed the economic growth rate beginning in fiscal year 2031.

When the interest rate on debt exceeds the growth rate of the economy that services it, the arithmetic turns. Before that line, a country can outgrow its burden if it can hold the primary deficit roughly flat. After that line, only primary surpluses bring the ratio down.

The United States has run primary surpluses in two of the last twenty-five years.

The mechanism of harm here is mechanical, not moral. It operates in three places.

• Crowding out. Every dollar going to interest is a dollar not going to anything else. Infrastructure. Education. Research and development. Defense. The investments that produce the growth that would, in turn, have lowered the ratio.
• Fiscal space. The 2008 response and the 2020 response were possible because the United States had room to borrow. The next crisis, whenever it arrives, will arrive at a debt-to-GDP ratio twice what the country carried into 2008. The room is narrower. In March 2026, a series of Treasury auctions saw weak demand, with primary dealers — the largest banks contractually obligated to absorb whatever the public does not buy — taking roughly twice the share they normally absorb of a two-year note. That is the kind of episode fiscal historians register — not as a forecast, but as one data point in a longer pattern of demand sensitivity that bears watching. Confidence in sovereign credit is not constant. It shifts.
• Sovereign debt crises are nonlinear. They do not arrive when a ratio crosses a threshold. They arrive when confidence breaks, and the mechanism, when it operates, operates fast. The United Kingdom in 2022, on a fiscal trajectory mild by American standards, broke its own gilt market within days. The United States is not the United Kingdom. The dollar is the reserve currency. The Treasury market is the deepest in the world. But the case that the United States is permanently exempt rests on conditions that themselves can change.

The case for treating the debt pattern as less urgent than it appears: the United States borrows in its own currency, controls its own central bank, and remains the safe asset of last resort. The 2008 crisis lowered American borrowing costs in the middle of an American-originated panic. The mechanism by which a fiscal crisis would arrive in an economy that retains those advantages is not obvious.

The counter-case is the one Capital and the Republic made most sharply in its Afterword.

• The danger is not the size of the figure.
• It is the habit of deferral that produced it.

The Grace Commission, 1984.

Bowles-Simpson, 2010.

The Joint Select Committee on Deficit Reduction — the supercommittee — 2011.

The fiscal cliff, 2013.

The debt-ceiling crises of 2011, 2013, 2023, and 2025.

The continuing resolutions that have, for most of the last fifteen years, replaced ordinary budget appropriations.

Each of these was a moment in which the political community recognized a structural fiscal difficulty, articulated it in detail, proposed responses to it, and then declined to enact them.

The deferral is not a series of accidents. It is the visible operating habit of a political community that has lost the capacity to make difficult fiscal decisions in real time.

That habit is the structural problem.

The deficit is the figure produced by the habit.

III. The Two Patterns, Considered Together

Now consider them together.

The wealth-concentration data and the debt data, set side by side, are not running in parallel.

They are running into each other.

The same fifty years that produced the rise of the top one percent’s share of household wealth from roughly twenty-two percent in the late 1970s to thirty-two in 2025 produced the rise of publicly held debt from twenty-five percent of GDP to ninety-nine. The same political community made both decisions. The decisions are not independent.

The mechanism that links them is the federal balance sheet.

When the political community of the post-1981 period chose to lower top marginal tax rates from seventy percent to twenty-eight (and, after partial reversals, to thirty-seven), to reduce the effective tax rate on capital relative to labor, and to leave estate taxation porous, it made a fiscal choice and a distributional choice in the same legislative breath.

The fiscal shortfall was financed by borrowing.

The borrowing was met by selling Treasury securities — to the same pool of accumulated private capital whose tax burden had been reduced.

The arithmetic compounds.

The top one percent now holds, directly and through pension funds and money-market vehicles, a substantial share of the federal debt it pays diminished taxes to service. The interest the federal government pays on that debt is income to the holders. The income to the holders, taxed at preferential rates on capital gains and dividends, returns less to the Treasury than it would have if collected as ordinary income tax on the wages it replaced.

The cycle closes.

This is the doom loop. Not the doom loop of bond mechanics, the one the credit-rating agencies watch. The doom loop of political economy.

• Wealth concentration produces political influence.
• Political influence produces tax and regulatory choices that raise the federal borrowing requirement.
• The borrowing requirement is met by selling claims on future taxation to the concentrated wealth.
• The claims on future taxation appreciate as a class of assets, deepening the concentration that produced the original political influence.

Each turn of the loop makes the next turn harder to interrupt.

There is a further consequence the trilogy’s framework lets us see clearly.

The levers for the two patterns constrain each other.

The inequality pattern, addressed alone, can be reached by the regulatory levers I named in Part II of the trilogy: financial regulation, antitrust enforcement, the modernization of the New Deal labor architecture for contemporary work. Those levers do not require new federal expenditure on a scale the debt trajectory now constrains.

The debt pattern, addressed alone, can be reached by the fiscal levers the deficit commissions have named, repeatedly: revenue increases, expenditure restraint, entitlement reform. Those levers require the political community to impose costs on identifiable constituencies in the present, against the resistance of organized interests, in exchange for benefits that accrue to a future political community that does not yet vote.

Treated separately, each set of levers is hard.

Treated together, they constrain each other.

Inequality reform that requires new expenditure runs into the debt constraint.

Debt reform that relies on revenue increases on capital runs into the political constraint produced by wealth concentration.

Debt reform that relies on expenditure cuts to programs serving the bottom half deepens inequality.

Inequality reform that relies on borrowing accelerates the debt trajectory.

There is no clean lever.

There is no single move.

The two patterns have become a single political problem with two faces.

IV. The Verdict

Which is worse?

I owe the reader a direct answer. I will give three, because the question has three.

• As immediate fiscal danger, the debt is worse. The line where the interest rate on debt exceeds the growth rate of the economy is now within five years on present projections. After thates. Wealth concentration, however severe, does not impose this kind of arithmetic constraint. The republic can carry thirty-two percent of wealth at the top for another decade without breaking. It cannot carry interest costs growing faster than tax revenue for another decade without breaking.
• As long-run threat to the republic, the wealth gap is worse. Debt, even mismanaged catastrophically, produces a crisis the political community can recognize, name, and respond to — as it did, on different terms, in 1947, in 1990, and in 2008. Inequality, once entrenched in the political system, captures the response. A polity that has lost the capacity to enact policy against the preferences of its top decile has lost something more fundamental than a balance-sheet position. It has lost the mechanism by which a republic governs itself. The arithmetic compounds against the country regardless of subsequent political choic
  • Insolvency can be cured by policy.
  • A hollowed-out democracy generally cannot.
• Treated together, the integration is worse than either alone. The deferral that has produced the debt is the same deferral that has produced the inequality. The political community that cannot bring itself to constrain capital cannot bring itself to tax it, either. The same habit underwrites both. The deferral on inequality is the parallel symptom of the deferral on debt.

The closest historical analogue is not 1929. It is the long stretch from 1873 to 1893 — the Gilded Age proper. A republic running two patterns at once, slowly, without an acute forcing condition, until something forced an answer.

The economic patterns of those twenty years — wealth concentration in Standard Oil and Carnegie Steel, an unregulated financial sector that produced the Panic of 1893, an industrial workforce without collective bargaining, a federal government too small to constrain any of it — were not corrected by the markets. They were corrected, eventually, by a generation of political reformers who built the Interstate Commerce Commission, the Sherman Antitrust Act, the income-tax amendment, the Federal Reserve, and the Pure Food and Drug Act. That reform sequence ran from 1887 to 1916. It was the work of three decades.

That period ended in the Progressive Era. The Progressive Era took twenty years to begin. The political community that produced it was preceded by the Pujo Committee’s documentation of financial concentration, by the Brandeis briefs, by the La Follette investigations, by the academic and institutional work of the 1920s. The five-year sprint of New Deal legislation from 1933 to 1938 was the visible portion of a far longer process.

The intellectual preparation for the present moment, if it is occurring, is occurring now.

V. What Comes Due

The pattern that comes due first is not, as the public debate assumes, the one that is larger.

It is the one that has been deferred longer.

Both have been deferred for fifty years.

Whichever breaks first will determine which we say was worse. We will not know in advance. We will know only afterward, in the way a republic always knows.

The data is on the table.

The levers are named.

The conditions are described.

What has been deferred is what comes due.

There is no soft answer to this.

The soft answers are what produced it.

*   *   *

Charles C. Jett, USNA ’64, is the author of Capital and the Republic, Debt and the Republic, and A Republic at Risk. A Naval Academy graduate and Cold War nuclear submarine officer whose 1969 tactical-paper innovations contributed to American undersea advantage during the Cold War, he holds an MBA from Harvard Business School and has taught at Harvard, Stanford, and Wharton. He hosts three podcasts — Making a Great America (a Federalist Papers series covering all 85 papers), Jefferson-Adams Letters, and It’s All About Skills — and writes at criticalskillsblog.com and civicsage.com.