A consumer economy that the world finances

Start with the composition, because it settles what kind of economy this is before any argument about policy. In the first quarter of 2026 output ran at an annual rate of 31,866 billion dollars. Personal consumption was about 68 percent of it. Private investment was about 18 percent, government consumption and investment about 17 percent, and net exports about minus 3 percent.

Read that as a sentence rather than a table. Roughly two thirds of American economic activity is households buying things, which means the single most important variable in the country is whether ordinary people feel able to spend. Investment and government are each a sixth, and trade subtracts.

The trade line is where most confusion lives, because the aggregate hides two opposite businesses. In calendar 2025 the country exported 3,430 billion dollars of goods and services and imported 4,362 billion, a deficit of 932 billion. But goods ran a deficit of 1,260 billion while services ran a surplus of 328 billion. The United States is a net importer of physical things and a net exporter of insurance, software, finance, travel, and licensing.

Who actually buys American goods is also narrower than the rhetoric suggests. Mexico took 337 billion dollars in 2025 and Canada 334 billion, together more than a fifth of goods exports and more than three times what went to China at 106 billion. The largest import sources are the same two countries, Mexico at 534 billion and Canada at 382 billion, ahead of China at 309 billion, Taiwan at 201 billion, and Vietnam at 194 billion. This is a continental trading bloc first and a global trader second.

And what it sells is less consumer-facing than most people expect. The largest goods export groups in 2025 were industrial supplies and materials at 784 billion dollars and capital goods excluding autos at 712 billion; consumer goods were 269 billion. At the single-commodity level the top lines were pharmaceutical preparations at 119 billion, crude oil at 99 billion, nonmonetary gold at 87 billion, civilian aircraft engines at 76 billion, and semiconductors at 68 billion.

So the honest one-line description is a consumption economy that exports inputs, machines, medicine, energy, and services, and imports the finished goods its households buy. Every question that follows in this dossier is downstream of that shape.

Three institutions do most of the work that shows up in prices, and it is worth knowing what each is actually instructed to do.

The central bank first. Congress created the Federal Reserve when Woodrow Wilson signed the Federal Reserve Act on 23 December 1913, and in 1977 it gave the institution the statutory objectives it still cites: to promote "maximum employment, stable prices, and moderate long-term interest rates." Two of those three are the famous dual mandate. The third is rarely mentioned and occasionally matters.

What stable prices means in practice is a number the Fed chose for itself rather than one Congress wrote. Its Statement on Longer-Run Goals says inflation at 2 percent, measured by the annual change in the price index for personal consumption expenditures, is most consistent over the longer run with its mandate. That statement was adopted in January 2012, substantively revised in August 2025, and most recently reaffirmed effective 27 January 2026. Notice what that means for reading any inflation print: the target is a specific index, chosen by the institution, and revisable by it.

Second, the fiscal machine, which is best understood through one fact. The last federal budget surplus was fiscal 2001, at about 128 billion dollars. Every year since has been a deficit, and fiscal 2025's was about 1,775 billion. Federal debt held by the public reached 98.7 percent of GDP in the first quarter of 2026, which is to say the government owes the public roughly one year of everything the country produces.

Third, and least visible, the arrangement by which that is financed. Foreign investors held 9,371 billion dollars of Treasury securities in May 2026, led by Japan at 1,143 billion, the United Kingdom at 949 billion, and mainland China at 659 billion. This is the country's most underappreciated institutional fact: a meaningful part of the price of American government borrowing, and therefore of American mortgages, is set by savers in other countries deciding what to do with their surpluses.

Underpinning all of it is the dollar's role, which is a privilege with a bill attached. The dollar was 57.1 percent of identified global foreign exchange reserves in the first quarter of 2026, against the euro at 20.0 percent, the yen at 5.4 percent, sterling at 4.4 percent, and the renminbi at 2.0 percent. One caveat that matters if you compare this to older figures: from the third quarter of 2025 the IMF stopped reporting an unallocated bucket and revised the series back to 2000, so current shares are of total world reserves and are not directly comparable to the older allocated-share numbers.

The regulatory posture has swung twice within living memory, and both swings are datable. The Gramm-Leach-Bliley Act of 12 November 1999 repealed the Glass-Steagall provisions separating commercial from investment banking. The Dodd-Frank Wall Street Reform and Consumer Protection Act of 21 July 2010 rebuilt a different set of constraints after the crisis. Any claim about American financial regulation being permanently loose or permanently tight is a claim about a variable that has reversed twice in twenty five years.

The resident population was 341.8 million on 1 July 2025. Be careful with anything more recent than that: the Census Bureau's own file carries short-term projections past that date, so a 2026 population figure is a projection rather than an estimate, and it should be labelled as one.

The median age was 39.4 years, and 64.6 million people, about 18.9 percent of the country, were 65 or over. Labour force participation was 61.5 percent in June 2026. Those three numbers together are the fiscal story of the next two decades: a rising share of the population drawing benefits, funded by a participation rate that has not returned to its pre-2020 level.

Then the number that governs the long run, and where the data has a gap worth knowing about. The total fertility rate was 1.60 in calendar 2024, a record low, well below the roughly 2.1 a population needs to replace itself without migration. There is no published 2025 total fertility rate: the National Center for Health Statistics dropped it from its 2025 provisional report, publishing a general fertility rate instead. So the most recent fact available is a year old, and anyone quoting a 2025 figure is computing it themselves or repeating someone who did.

Which makes migration the swing variable rather than a side issue. Net international migration was about 1.26 million in the year to July 2025. With fertility at 1.60, migration is not a supplement to American population growth. It is close to the whole of it, which is why an argument about immigration policy is also, whether the participants say so or not, an argument about the future size of the labour force and the tax base.

The scale of the whole arrangement is worth one backward glance. The first census in 1790 counted 3,929,214 people. The 2020 census counted 331,449,281. The country has grown roughly eighty four fold in two hundred and thirty years, and almost none of the mechanism that produced that growth is currently operating: fertility is below replacement and the frontier is closed.

The clearest way into the American financial system is to ask what the government pays to borrow, because almost every other price in the country is quoted off it.

On 23 July 2026 the ten year Treasury yielded 4.71 percent and the two year 4.37 percent, so the curve carried a positive slope of about 34 basis points. The thirty year fixed mortgage averaged 6.58 percent that week. Those three numbers are the transmission mechanism: the policy rate anchors the short end, the long end is set by a global market including the foreign holders of the previous section, and the mortgage sits on top of the long end plus a spread.

The banking system is the other half. Total bank credit was 19,741 billion dollars in mid July 2026, growing about 6.4 percent over the year, and commercial and industrial loans were 2,894 billion in June, up about 8.0 percent. Delinquency on all loans was 1.48 percent in the first quarter, against a peak of 7.40 percent in early 2010, and the net share of banks reporting tighter standards on commercial and industrial loans was 8.1 percent in the April 2026 survey.

Read together, those are the readings of a system that is neither stressed nor euphoric. Credit is growing faster than nominal output, standards are mildly tight, and losses are low. Whether that is a comfortable equilibrium or the pleasant part of a cycle is exactly the question the credit cycle brief exists to make answerable, and it is worth reading beside this section rather than in place of it.

The currency deserves its own paragraph, because the dollar's reserve role is what allows the arrangement described in this dossier to persist. A country running a trade deficit near a trillion dollars must be receiving a matching inflow of capital, by accounting identity. The dollar's 57 percent share of global reserves, and the 9.4 trillion dollars of Treasuries held abroad, are what that inflow looks like in practice. Foreign savers accumulate dollars from selling goods to Americans, and they put a large share of them back into American government debt.

So the deficit and the financing are not two facts, they are one fact told from two ends. That has a sharp implication worth holding: the arrangement is stable as long as foreign savers want dollar assets, and the question to ask about it is never whether the deficit is large but whether anything is changing about that willingness. The renminbi at 2.0 percent of reserves is the usual candidate for a challenger, and at that level it is a candidate rather than a rival.

Four dates carry most of what a reader needs, and each one is a moment when the country chose the arrangement it still lives in.

1913. Congress created a central bank, and Woodrow Wilson signed the Federal Reserve Act on 23 December. Before that the United States had spent most of its history without one, and the panics that produced the Act are the reason the institution exists at all.

1944. Delegates from forty four nations met at Bretton Woods, New Hampshire, in July, and created the International Monetary Fund and what became the World Bank. The system they built fixed other currencies to a dollar that was itself convertible into gold, which is how the dollar acquired the central role it still holds.

1971. On 15 August, Richard Nixon told the country he had "directed Secretary Connally to suspend temporarily the convertibility of the dollar into gold or other reserve assets." The word temporarily has now stood for over half a century. What replaced Bretton Woods was a dollar backed by nothing but the American economy and the depth of its markets, and every subsequent argument about reserve currencies dates from that address.

1977. Congress wrote the Federal Reserve's objectives into statute: maximum employment, stable prices, and moderate long-term interest rates. The 2 percent inflation target is not in that statute. The Fed adopted it in 2012 and has revised the surrounding strategy since, most substantively in August 2025.

Two more dates matter for how the financial system is shaped, and they point in opposite directions. In November 1999 Gramm-Leach-Bliley repealed the Glass-Steagall separation of commercial and investment banking. In July 2010 Dodd-Frank built a new regulatory architecture after the crisis that followed. The lesson is not that either was right. It is that the American financial system's rules are revisable on a timescale of roughly a decade, which is shorter than the life of most of the assets they govern.

Set against those institutional dates, one economic fact for scale. The country produced about 26 percent of world output measured at market exchange rates in 2025, and about 14.6 percent measured at purchasing power parity. Both figures are correct and they answer different questions: the first is how much the rest of the world will pay for what America makes, the second is how much America actually makes. Anyone using one to argue about the other is comparing two different things.

As of late July 2026 the readings do not line up into a single story, and the honest thing to do is say what each one says and name what would settle the disagreement.

Growth has been erratic rather than directional. Real output grew at an annual rate of 4.4 percent in the third quarter of 2025, then 0.5 percent in the fourth, then 2.1 percent in the first quarter of 2026. A three quarter run like that is not a trend, and averaging it into one number discards the information that the volatility itself carries.

The labour market is cooling on every measure at once, which is more informative than any one of them. Unemployment was 4.2 percent in June 2026, up from a cycle low of 3.4 percent in April 2023. Payrolls added 57,000 jobs in June, and the three month average was about 111,000. Job openings were 7.59 million in May. None of those is a recession reading; all of them are moving the same way.

Inflation is the awkward part, and it is awkward in an unusual direction. Consumer prices rose 3.46 percent over the year to June 2026 with core at 2.57 percent, so the headline is running above the core. On the Fed's preferred measure the gap is wider still: personal consumption prices rose 4.07 percent over the year to May with core at 3.41 percent. Headline above core means the pressure is coming from the volatile components, energy and food, rather than from the broad wage-and-services engine, and the June statement pointed at energy supply.

That distinction decides how a central bank should respond, which is why it is worth being careful about. Core above headline suggests entrenched inflation and argues for tightening. Headline above core suggests a supply shock passing through, which monetary policy can do little about and which will fall out of the annual comparison on its own. The Fed's own posture reads as the second interpretation: it cut in December 2025 and has held the target range at 3.50 to 3.75 percent since, most recently on 17 June 2026.

The credit side is quiet, which is the piece that argues against reading all of the above as late cycle. Bank credit is growing about 6.4 percent, standards are only mildly tight at a net 8.1 percent tightening, mortgage standards are net easing, the Baa spread over Treasuries is 1.58 points, and delinquencies are 1.48 percent. Nothing in the credit data looks like 2007 or even like 2023.

So the two readings to hold at once. Cooling labour market plus supply-driven inflation plus quiet credit is a soft landing in progress. Cooling labour market plus a Fed that has stopped cutting plus a curve at only 34 basis points of slope is the front half of something worse. What separates them is not more opinion, it is which series turns next: if the credit indicators tighten while payrolls keep slowing, the second reading is winning; if headline inflation falls back toward core while payrolls stabilise, the first is.

One last practical warning about the data itself. October 2025 is missing outright from both the consumer price index and the unemployment series. Any twelve month comparison spanning that month is built over a hole, and anyone who hands you a clean year-over-year figure through it has interpolated something without telling you.