Stress-Testing the American Empire: Real Decline or Cyclical Strain?

In 1942, with German submarines savaging British Atlantic convoys and Rommel pressing into Egypt, the editors of The Times of London looked past the war and wrote that their empire, still covering a quarter of the globe, had already become "a self-liquidating concern." Eighty-four years later, in May 2026, New York Times contributing editor Christopher Caldwell published a column under the headline "America Is Officially an Empire in Decline," arguing that Donald Trump's intervention in Iran had become "a watershed in the decline of the American empire." Both passages survive in a widely syndicated essay by historian Alfred W. McCoy, carried this year by Asia Times and Fair Observer, which places the two editorials side by side and asks whether Washington is replaying London's script.

The parallel is rhetorically powerful. It is not, by itself, evidence. Prediction of imperial decline is an old American genre; Sputnik, Vietnam, and the Japanese industrial surge each produced bestsellers in it, and the republic outlasted every one of those verdicts. What makes the current round worth taking seriously is measurable: the arithmetic of federal debt, the behavior of the $28-plus-trillion Treasury market, the composition of the world's currency reserves, the procurement decisions of allied governments, and the throughput of American shipyards. These series exist. They can be read.

This article reads them. The stakes are not abstract. The dollar's reserve role underwrites mortgage rates in Ohio and sanction regimes in the Persian Gulf; the American alliance umbrella determines whether 50-odd treaty partners spend their own money on defense or spend it hedging; the fiscal trajectory decides how much capacity Washington has left when the next crisis arrives. The question below is narrower than whether America is "in decline." It is whether today's strain is cyclical, meaning painful but reversible, or structural, meaning compounding regardless of who governs. The answer, the evidence suggests, is that it depends which pillar of the system you inspect, and that four of them are worth watching on a timer.

What counts as "empire," and what counts as "decline"?

The United States is not an empire of governors-general and colonial offices. Its imperial position is infrastructural: roughly 750 overseas military installations in more than 80 countries (the public mapping built by anthropologist David Vine is the standard reference), a web of collective-defense treaties binding America to most of Europe, Japan, South Korea, and Australia, the world's dominant reserve and transaction currency, outsized sway over the IMF, World Bank, and sanctions machinery, and a security-patronage role that made American firms responsible for about 43 percent of global arms exports over 2020-2024, according to SIPRI's arms-transfer accounting.

"Decline" also needs a working definition, because two very different claims get conflated. Relative decline means America's share of world economic weight shrinks as others grow. Absolute decline means Washington loses the ability to perform hegemonic functions: financing its commitments, setting rules others follow, deterring rivals, and commanding voluntary followership. A 2025 study in Frontiers in Political Science (Patel, Hansmeyer, Desai, and Ajit) defines a functioning hegemon by three attributes: material capacity to shape the rules, the will to lead, and "consented followership." All three are testable. Relative decline, on the data, is established fact. Whether it has tipped into absolute decline of function is what the rest of this article examines.

Does the fiscal base still support the commitments?

History's clearest marker of imperial overextension is when a great power spends more servicing its debts than defending its frontiers. Historian Niall Ferguson has called this "Ferguson's Law," citing Habsburg Spain, pre-revolutionary France, the Ottomans, and late Britain.

The United States crossed that line in fiscal 2024. Net interest on the federal debt came to roughly $880 billion, according to Treasury's Monthly Statement, against a defense budget near $850 billion. Gross interest outlays passed $1 trillion a year during 2024-2025. The Congressional Budget Office's Long-Term Budget Outlook for 2025-2055 projects debt held by the public at about 100 percent of GDP in 2025, surpassing the World War II record of 106 percent before the decade ends, and reaching 156 percent by 2055, with deficits staying near or above 6 percent of GDP for most of the projection window and passing 7 percent by mid-century, sustained territory the United States has historically reached during wars and recession aftermaths, not in peacetime expansion. In July 2025, CBO estimated the One Big Beautiful Bill Act would add about $3.4 trillion to deficits over a decade, more if its temporary provisions are made permanent.

Ratings agencies have registered the trajectory in sequence: S&P in 2011, Fitch in 2023, and on May 16, 2025, Moody's stripped the United States of its last top-tier AAA rating. What makes this structural rather than cyclical is the compounding: interest is the one budget line no Congress votes on annually, and each refinancing at higher yields enlarges it. The Treasury market itself has shown intermittent fragility, including the April 2025 turbulence around tariff announcements and persistent concern about dealer capacity and hedge-fund basis-trade concentration. Writing on September 1, 2026, economist Jack Rasmus argued that stress in the Treasury market was again building in recent weeks; that is commentary, not confirmation, but it reflects how crowded the fiscal-stress watch has become.

Are creditors hedging the dollar?

The dollar remains the world's currency by wide margins: it sits on one side of 88 percent of all foreign-exchange trades (BIS 2022 triennial survey) and accounts for roughly half of international payments by value in SWIFT data. But the reserve data describe a slow, two-decade leak. IMF COFER statistics put the dollar's share of disclosed global reserves at about 57.8 percent at the end of 2024, down from around 71 percent at the turn of the century. A 2022 IMF working paper by Arslanalp, Eichengreen, and Simpson-Bell found the decline is real even after stripping out valuation effects, and is flowing mostly into nontraditional reserve currencies rather than any single rival.

Gold is the louder tell. Central banks bought more than 1,000 tonnes of it in 2022, 2023, and 2024, the longest such streak in decades, according to the World Gold Council's quarterly demand reports. The European Central Bank's 2025 International Role of the Euro report calculated that gold, at market prices, had overtaken the euro as the second-largest reserve asset globally. China's own Treasury holdings have fallen from a 2013 peak of about $1.3 trillion to below $800 billion in Treasury TIC data, while its central bank adds gold and expands the yuan-based CIPS payments network. The 2022 freezing of Russian central-bank reserves hangs over all of this; reserve managers noticed.

Read together, the verdict is narrow: creditors are diversifying at a pace of roughly half a percentage point of share per year, not fleeing. The diagnostic is the pace, not the direction. A reserve transition that took Britain decades could compress dramatically in a modern crisis, which is why the quarterly series matter more than any single snapshot.

Can the military carry the strategy?

Measured in money, American primacy is intact. Washington spent about $997 billion on defense in 2024, 37 percent of the world total and more than three times China's $314 billion estimate, per SIPRI's military-expenditure database. It fields 11 nuclear-powered aircraft carriers to China's three conventionally powered ones, operates the only global network of overseas bases, and provides the refueling, lift, and intelligence enablers no ally can yet replicate. Its defense burden, near 3.4 percent of GDP, is modest by Cold War standards.

The strain is industrial, not budgetary. A 2023 Navy briefing slide reported by USNI News compared China's shipbuilding capacity, roughly 23 million tons, to America's less than 100,000, a gap on the order of 230-to-1. The fleet hovers around 296 battle-force ships against statutory and Navy goals of 355 to 381; Virginia-class submarines are being delivered at about 1.2 boats a year against a target of two; the Army's race to expand 155mm artillery output from about 14,000 rounds a month before 2022 toward a 100,000-a-month goal has slipped toward 2026. The Army missed recruiting targets in 2022 and 2023 before recovering. The bipartisan Commission on the National Defense Strategy concluded in July 2024 that the United States is not prepared for a major war across multiple theaters, and CSIS wargames have long flagged munitions exhaustion in any Taiwan scenario.

June 2025 illustrated both faces of the ledger. The strike on Iran's nuclear sites, B-2s flying from Missouri and expending a reported 14 GBU-57 bunker-busters, a large share of a stockpile believed to number only in the low dozens, demonstrated reach nobody else possesses and a munitions depth nobody should confuse with abundance. When Caldwell wrote eleven months later, as quoted in McCoy's essay, that Trump "has overextended the empire dangerously," he was naming, in effect, this gap between strategy and means, the most concrete sense in which "overstretch" exists today.

Are allies rearming to help Washington or to hedge against it?

Both, and the mix is the story. When Russia invaded Ukraine, just three NATO members were spending 2 percent of GDP on defense as recently as 2014; by 2024, twenty-three of thirty-two met that floor. At The Hague in June 2025, the alliance pledged to reach 5 percent of GDP by 2035, 3.5 percent on core military spending plus 1.5 percent on related infrastructure. That is burden-sharing Washington demanded for decades.

Yet the procurement signals carry insurance logic. The European Commission's March 2025 rearmament program (ReArm Europe, later Readiness 2030) aims to mobilize up to €800 billion with eligibility rules that favor European suppliers. In March 2025, Canada's new government ordered a review of its F-35 purchase, Portugal's defense minister publicly questioned the aircraft over dependence on American logistics, and Spain expanded Eurofighter orders rather than turning to the F-35. Germany's chancellor has floated consulting France and Britain about a European nuclear dimension. SIPRI's transfer data cut in both directions: European states' arms imports surged 155 percent over 2020-2024, with about 64 percent coming from the United States. Allies are more dependent on American kit in the near term, and more determined to reduce that dependence by design. That is hedging with a purchase order attached.

Are rivals building an exit ramp?

China's hedging portfolio is the most systematic of any actor: a $500-plus-billion reduction in Treasury holdings from the peak, steady gold accumulation, yuan settlement arrangements that Russian officials put at more than 90 percent of bilateral trade, and a decade of industrial policy aimed at chip self-sufficiency, which produced a 7-nanometer smartphone processor in 2023 despite export controls. Supply chains are being physically moved around both governments. China's share of U.S. imports has fallen from about 21 percent in 2017 to roughly 14 percent in 2024, Census data show, with Mexico now America's top trading partner and Vietnam absorbing assembly work, though much of that rerouted output still contains Chinese content. Washington's 2025 tariff rounds pushed partners to diversify further, from EU trade deals with South America to Asian economies deepening intra-regional trade inside RCEP. None of this replaces the American market or the dollar. All of it builds optionality, which is what hedging means.

Where does America still widen its lead?

The strongest rebuttal to the decline thesis is not in budgets or bases; it is in innovation and resources. Stanford's 2025 AI Index found U.S. private AI investment hit $109.1 billion in 2024 against China's $9.3 billion, and that American institutions produced 40 notable AI models to China's 15 and Europe's three; the United States also remains the leading workplace for elite AI researchers, per MacroPolo's talent tracker. American firms dominate chip design, design software, and semiconductor equipment even though leading-edge fabrication sits in Taiwan; a 2024 SIA/BCG study projects the United States could reach about 28 percent of leading-edge logic capacity by 2032 as fab construction from the $52.7 billion CHIPS Act comes online. American crude production hit a record average of about 13.2 million barrels a day in 2024, per the EIA, alongside the world's largest LNG export position, and the country hosts nearly two-thirds of the MSCI global equity benchmark's value. Demography, the quiet variable, also favors America: China's population has been shrinking since 2022 and its working-age cohort since the middle of the last decade, while U.S. population keeps growing, mostly through immigration, which the current crackdown puts at some risk. A Berkeley Lab study for the Energy Department warns the constraint is shifting to power, projecting data centers rising from about 4.4 percent of U.S. electricity use in 2023 to anywhere from 6.7 to 12 percent by 2028. The edge is real; its bottleneck is grid capacity, not inventiveness.

Is the factory floor coming back?

The hollowed-out heartland is where the decline narrative feels most like lived experience, and the data justify the feeling selectively. Manufacturing contributes about 10 percent of U.S. GDP (BEA) and employs roughly 13 million people, about 8 percent of the workforce, down from a 1979 peak near 19.6 million; China now accounts for roughly 30 percent of global manufacturing value added to America's 16 percent, on World Bank figures. The U.S. goods trade deficit hit a record $1.2 trillion in 2024. Against that, factory construction spending roughly tripled between 2021 and late 2024, to about a $235 billion annual rate, the most intense build-out since the record-keeping began, driven by semiconductors, batteries, and the CHIPS Act. What has not yet followed is the jobs: manufacturing employment was essentially flat through 2024 and 2025, and the payoff from the construction boom depends on fabs actually ramping, from TSMC in Arizona to the next tranches of planned plants. Reshoring policy has bought capacity. Whether it bought production is this decade's open question.

Is Washington losing the capacity to govern?

The least cyclical problem may be the state itself. Pew's long-running series found just 22 percent of Americans trusting the federal government to do the right thing most of the time in 2024, against 77 percent in 1964. Congress has not enacted all appropriations bills on time since the mid-1990s, governing instead through continuing resolutions and debt-limit brinkmanship, politics that S&P cited outright in 2011, that Fitch echoed in its 2023 downgrade, that Moody's flagged in its 2023 outlook warning, and that formed the backdrop to Moody's 2025 downgrade, which rested mostly on the fiscal math itself. The IRS projected an annual tax gap approaching $700 billion for tax year 2022, revenue owed but uncollected, after a decade of enforcement capacity erosion, partial rebuilding under the 2022 Inflation Reduction Act, and then 2025's reversal of much of that rebuilding as part of the administration's cost-cutting drive. Since January 2025, the executive branch has fired inspectors general en masse, terminated tens of thousands of probationary federal employees in actions later curtailed by courts, moved to convert policy-adjacent career positions into political appointments via executive order, and frozen or clawed back congressionally appropriated research funds at universities and science agencies. Government can lose competence faster than it can lose wars, and this is the decline marker most under Washington's own control.

What do Rome, Britain, and the Soviet Union actually say?

Rome's relevance is mostly negative: an agrarian empire without bond markets teaches little directly beyond the fiscal logic that debasement eventually catches up with frontier costs. Britain is the instructive case. The 1942 editors were right within five years; the sterling crisis at Suez in 1956 revealed, in a single episode, that a bankrupt metropole could not defy its creditor; and yet decline arrived as a managed, decades-long handoff rather than a fall, with reserve-currency status decaying slowly across the postwar era and London winding up "on reasonably good terms," as Caldwell notes, with most former colonies. The Soviet case teaches the opposite speed limit: a system whose burden economists variously estimated at 15 to 17 percent of GDP for defense went from superpower to dissolution in roughly five years, and outside analysts, including the CIA, misread its strength until nearly the end. McCoy's essay argues modern empires live shorter lives, citing Britain's and the Soviet Union's spans to suggest America's 80 years already exhaust the allotment. That is an argument, not a law.

The quantitative literature lands in the middle. The Frontiers study models historical hegemon lifecycles and projects potential turning points for American primacy between 2032 and 2067, while noting the U.S. already exhibits recurring decline markers: internal division, weakened alliances, eroding legitimacy. World-systems researchers at UC Riverside, extending measures tracked since 2005, describe a slow, grinding decline in U.S. relative economic and political weight over recent decades, not a collapse curve. And the revivalist school, from Paul Kennedy's 1987 overstretch thesis (which predated a decade of American expansion) through Josef Joffe's critique of "declinism" to Michael Beckley's argument that China is peaking, offers a standing warning: every prior decline call failed on America's capacity for reinvention. The honest reading of history is that systems held together by confidence look stable until they re-rate quickly, and the people best positioned to see it coming are usually the ones already living it, as those Times editors were.

So: decline, strain, or something in between?

The evidence, weighed as of September 2026, supports a precise verdict. Relative economic decline is established fact: IMF purchasing-power estimates for 2026, cited in McCoy's essay, put China near 20 percent of world output, the United States near 15, and the EU around 14, while on market-exchange terms the U.S. share has held near 26 percent for decades. The decline that is structural rather than cyclical concentrates in two places: the fiscal trajectory, which compounds mechanically and has survived every administration of both parties this century, and state capacity, which is being eroded deliberately. The pillars that remain strained but dominant, dollar, military reach, technology, energy, capital markets, show hedging at the margins without an alternative at the center.

That makes the assignment's central hypothesis worth stating carefully. The diagnostic signal historians associate with late empire is simultaneous hedging by creditors, allies, and rivals at once, and that triangulation is now visible: creditors drift from dollar reserves into gold, allies write autonomy into procurement policy, and rivals build payment and supply-chain exit ramps. But "visible" is not "advanced." Each hedge is shallow, reversible, and individually old. What would convert the pattern from cyclical strain to structural decline is acceleration arriving on multiple channels together. On the modeling range from the Frontiers paper, and on CBO's arithmetic alike, the 2030s are the test decade; the current one is the window in which the outcome is still being chosen. And the decline thesis has falsifiable rivals: if primary deficits narrow, if the factory-construction boom converts into output share, or if allies' new money builds complementing rather than substituting capabilities, the strain was a cycle. Watch the data, not the editorials.

What should you watch every quarter?

The dashboard below separates structural signal from headline noise. Readings reflect the most recent official releases cited in this article; several series lag by a quarter or more.

If several cells in the right-hand column fill at once, the question this article opened with will have answered itself, quietly, in tables rather than editorials.

Sources and references

1. Alfred W. McCoy, "An Autopsy of American Empire," Asia Times, June 2026; syndicated as "After America: The Causes and Consequences of US Global Decline," Fair Observer, August 26, 2026. https://asiatimes.com/2026/06/an-autopsy-of-american-empire/ and https://www.fairobserver.com/politics/after-america-the-causes-and-consequences-of-us-global-decline/
2. Jack Rasmus, "The Decline of American Empire Part 1: The Funding Gap," September 1, 2026. https://jackrasmus.com/2026/09/01/the-decline-of-american-empire-part-1-the-funding-gap/
3. Ketan Patel, Christian Hansmeyer, Nandan Desai, Aditya Ajit, "American hegemony at a critical juncture: lessons from history's great powers," Frontiers in Political Science, September 2025. https://doi.org/10.3389/fpos.2025.1511913
4. Christopher Chase-Dunn and Emlyn Zhai, "The Rise and Decline of U.S. Hegemony in Evolutionary Perspective," IROWS Working Paper 147, University of California-Riverside, April 2025. https://irows.ucr.edu/papers/irows147/irows147.pdf
5. Congressional Budget Office, "The Long-Term Budget Outlook: 2025 to 2055," March 2025, and "The Budget and Economic Outlook: 2025 to 2035," January 2025. https://www.cbo.gov/
6. Committee for a Responsible Federal Budget, analysis of the One Big Beautiful Bill Act, July 2025. https://www.crfb.org/
7. U.S. Treasury, Monthly Treasury Statement, fiscal year 2024 receipts and outlays. https://fiscal.treasury.gov/reports-statements/mts/
8. Moody's Ratings, downgrade of the United States to Aa1, May 16, 2025. https://www.moodys.com/
9. IMF, Currency Composition of Official Foreign Exchange Reserves (COFER). https://data.imf.org/
10. Serkan Arslanalp, Barry Eichengreen, Chima Simpson-Bell, "The Stealth Erosion of Dollar Dominance and the Rise of Nontraditional Reserve Currencies," IMF Working Paper 2022/058. https://www.imf.org/
11. Bank for International Settlements, Triennial Central Bank Survey of Foreign Exchange and OTC Derivatives Markets, 2022. https://www.bis.org/
12. World Gold Council, Gold Demand Trends, quarterly reports 2022-2025. https://www.gold.org/goldhub/research/gold-demand-trends
13. European Central Bank, "The International Role of the Euro," June 2025. https://www.ecb.europa.eu/
14. U.S. Treasury, Treasury International Capital data, Major Foreign Holders of Treasury Securities. https://ticdata.treasury.gov/
15. SIPRI, "Global military spending reaches historic highs" (2024 data), April 2025; "Trends in International Arms Transfers, 2020-24," March 2025. https://www.sipri.org/
16. NATO, The Hague Summit Declaration, June 25, 2025, and annual defense-expenditure estimates. https://www.nato.int/
17. European Commission, ReArm Europe / Readiness 2030 communication, March 2025. https://commission.europa.eu/
18. Congressional Research Service, "Navy Force Structure and Shipbuilding Plans: Background and Issues for Congress" (RL32665), and defense-industrial-base reports. https://crsreports.congress.gov/
19. Commission on the National Defense Strategy, Final Report, July 2024 (hosted by the RAND Corporation). https://www.rand.org/
20. Seth G. Jones, "Empty Bins in a Wartime Environment: The Challenge to the U.S. Defense Industrial Base," CSIS, 2023. https://www.csis.org/
21. USNI News, reporting on Office of Naval Intelligence comparison of U.S. and Chinese shipbuilding capacity, July 2023. https://news.usni.org/
22. U.S. Census Bureau, Manufacturing Construction Spending (C30) and U.S. International Trade in Goods and Services. https://www.census.gov/
23. Bureau of Economic Analysis, GDP by Industry. https://www.bea.gov/
24. Bureau of Labor Statistics, Current Employment Statistics, manufacturing series. https://www.bls.gov/
25. World Bank, World Development Indicators, manufacturing value added. https://data.worldbank.org/
26. Stanford Institute for Human-Centered AI, "The 2025 AI Index Report." https://hai.stanford.edu/ai-index
27. MacroPolo (Paulson Institute), The Global AI Talent Tracker. https://macropolo.org/analysis/the-global-ai-talent-tracker/
28. Semiconductor Industry Association and Boston Consulting Group, "Emerging Resilience in the Semiconductor Supply Chains," May 2024. https://www.semiconductors.org/
29. U.S. Energy Information Administration, crude oil production data. https://www.eia.gov/
30. Lawrence Berkeley National Laboratory for the U.S. Department of Energy, "2024 United States Data Center Energy Usage Report," December 2024. https://www.lbl.gov/
31. Pew Research Center, "Public Trust in Government: 1958-2024," June 2024. https://www.pewresearch.org/
32. Internal Revenue Service, tax gap projections for tax year 2022, October 2024. https://www.irs.gov/
33. Executive Order 14171, "Restoring Accountability to Policy-Influencing Positions Within the Federal Workforce," January 20, 2025, Federal Register. https://www.federalregister.gov/
34. U.S. Government Accountability Office, High-Risk List. https://www.gao.gov/high-risk-list
35. Federal Reserve Bank of New York, ACM Term Premia data. https://www.newyorkfed.org/research/data_indicators/term-premia-tabs
36. Federal Reserve Bank of St. Louis, FRED series FYFSGDA188S (Federal net outlays: interest as percent of GDP). https://fred.stlouisfed.org/
37. David Vine, "Base Nation" and the American University overseas bases mapping project. https://davidvine.net/
38. Paul Kennedy, "The Rise and Fall of the Great Powers" (Random House, 1987); Josef Joffe, "The Myth of America's Decline" (Liveright, 2014); Michael Beckley, "Unrivaled: Why America Will Remain the World's Sole Superpower" (Cornell University Press, 2018).
39. Niall Ferguson, 2024 commentary proposing "Ferguson's Law" on debt service versus defense spending.
40. SWIFT, RMB Tracker and payments-by-currency statistics (U.S. dollar share of global payments by value). https://www.swift.com/