Trail 2026-10-11 (b, second session)

Budget: 15 fetched pages or 40 tool calls. Orientation: steer, index, research-values, dead-ends, trail-2026-10-11, macro-view, thread-transition.

Thread chosen: macro-view (within thread-transition), because steer item 1 says so: test the four crises (dollar debt, resources, fossil/climate, deglobalisation) separately, evidence for and against. Aim: one primary, measurable indicator per crisis, plus whatever bends the "age of crisis, not a crisis" framing.

Hops

Tooling: web_extract still fails (search-only backend). The free-tier search fallback returned long text from primary pages; pages read in full went through the browser. Each hop says which.

1. Expect: DHL's connectedness data, the most direct measure of deglobalisation; could knock out a leg. -> https://www.dhl.com/content/dam/dhl/global/microsites/dhl-nyu/pdf/dhl-global-connectedness-tracker-mar-2025.pdf (search text): depth at a record 25% since 2022; "Deglobalization remains a risk, not a current reality." Strongly against.
2. Expect: IMF COFER, the primary series on the dollar's reserve share. -> https://data.imf.org/en/news/imf%20data%20brief%20march%2027 : 56.77% in 2025Q4; the Q2 2025 fall was almost all FX valuation. Against "dollar crisis" in the reserve sense.
3. Expect: the academic polycrisis paper, to see whether anyone has given "age of crisis" a mechanism. -> Cambridge page, read in the browser: triggers + slow stresses; three pathways; hyper-connectivity as the transmission medium. Bears on the frame; pulls against the deglobalisation leg.
4. Expect: whether the 2026 Hormuz shock changed DHL's verdict. -> DHL Globalization Tracker Oct 2026 press release (prnewswire): record globalisation; H1 2026 trade growth fastest in 15 yrs; AI goods 76% of Q1 trade growth; Saudi -37%. Surprise: the war barely registered globally.
5. Expect: CBO's own figures for the debt leg. -> https://www.cbo.gov/publication/61882 : interest above 3.2% of GDP every year to 2036, highest since 1940; debt 120% by 2036. For.
6. Expect: Global Carbon Budget primary numbers. -> record fossil CO2 in 2025, 38.1 Gt; ~4 yrs of 1.5°C budget left. For.
7. Expect: Arrighi's signal/terminal crisis (old and forgotten; crosses debt and hegemony). -> found Raj Patel's essay of 15 Sep 2026, read in full in the browser. It maps the Treasury safety premium and finance's profit share onto Arrighi: terminal crisis since 2008. Surprise: someone has already plotted James's dollar-debt leg as an Arrighian cycle, using Lustig's data.
8. Expect: Lustig's primary paper. -> https://www.economicstrategygroup.org/wp-content/uploads/2026/08/Lustig-2.pdf (long search text): Treasury convenience yield gone; stock-bond hedge broken; implicit fiscal dominance loop. For, and more specific than "dollar debt".
9. Expect: IEA on critical-minerals concentration. -> top-3 refining 82->86%; top refiner a record 70% in 2025; copper gap of 30% by 2035; rare earths the exception. For, as concentration rather than depletion.
10. Expect: Ember on fossil power, as a counter. -> fossil generation -0.2% in 2025; wind+solar out-generated gas in April 2026 during Hormuz, with no switch to coal. Against the fossil leg, for the climate leg.
11. Expect: a test of "the dollar is no longer a haven" in the Iran war. -> AEI piece blocked by Cloudflare; the advisoranalyst page came back empty. Headlines only (Reuters, CNBC: dollar rallied in early March). Unresolved; a dead end for now.
12. Expect: UNCTAD WIR 2026 as the counterweight to DHL on composition. -> strategic sectors 16->44% of greenfield; non-strategic manufacturing -17%, low-income -70%; SOEs 26 of top 100. Bends the deglobalisation leg into "securitised globalisation".

Stopping rule not triggered. Stopped with budget left, to write up.

Hunches (unsourced, not in wiki)

• The four legs may share one driver that is not on James's list: the AI capex boom. It drives the trade record (DHL) and the greenfield concentration (UNCTAD), adds data-centre power demand that blunts the fossil peak, and plausibly feeds part of the deficit/bond story. Unsourced as a unifying claim.
• "Securitised globalisation" may be a better name for the fourth leg than "deglobalisation". My phrase, not a source's.

End of session

Surprises

• The 2026 Hormuz closure barely dented global trade. H1 2026 goods trade grew faster than in any half-year of the past 15 years bar Covid, with AI hardware 76% of Q1 growth. Deglobalisation, as stated, is the weakest of the four legs. https://prnewswire.com/news-releases/tech-beats-tariffs-global-trade-growth-accelerates-and-globalization-sets-new-record-302900608.html
• The dollar leg holds, but not where expected. The reserve share is flat (~57%) while the Treasury safety premium has vanished, and Raj Patel has already plotted this as Arrighi's terminal crisis. https://rajpatel.org/2026/09/15/the-rise-and-fall-of-americas-exorbitant-privilege/
• During the oil and gas shock, wind+solar out-generated gas worldwide for the first month ever (April 2026). https://ember-energy.org/latest-updates/for-the-first-time-wind-and-solar-generated-more-electricity-than-gas-worldwide-in-april-2026/

Dead end: whether the dollar and Treasuries acted as a haven in the March 2026 war. AEI was blocked by Cloudflare and the other piece was empty. Worth continuing: macro-view, deglobalisation leg. Next hop: the DHL Oct 2026 tracker PDF itself (share of trade between geopolitical blocs, 2024-26), to see whether bloc fragmentation moved after Hormuz. Permission to chase: the postscript to the 2009 edition of Arrighi's The Long Twentieth Century, as a primary source for the "age of crisis" frame. It would cross threads 2 and 1 (who owns infrastructure at the end of a cycle). Budget used: 12 web fetches/searches (of 15); ~24 tool calls (of 40).