Anthony Bardaro @AnthPB

…re my aforementioned "funding gap"¹ – here's the latest on AI capex and financing (sycamorelp.com/wp-content/upl…): GLOBAL DATACENTER CAPEX: $2.9T thru 2028... ➖ $1.4T HCP cashflows (42%) 🟰 $1.5T funding gap (52%) ➕ $300B energy (excluded) AI DEBT ISSUANCE: $500B 2026YTD annualized² IG TECH CREDIT: 0.8x net leverage (as reported) 1.8~2.0x including leases (+$932B³) 1.8x for non-financial IG composite 2.5–3.0x unused debt capacity (i.e. credit downgrade) HCP FCF⁴: 2024: $235B (peak) 2025: $230B 2026/27e: $75–95B (trough) 2030e: ~$470B POWER BOTTLENECK (US): 68 GW need thru 2028 ➖ 15 GW under construction ➖ 15 GW available/contracted grid interconnection 🟰 38~45 GW shortfall ➖ remedies⁵ 🟰 -16% deficit TELECOM ANALOGUE: 3-4 years until defaults peaked (i.e. 2028-30 after 2025-26 AI issuance boom) 56% of 2002 defaults by dollar volume 31% " " " by issuer count 23% recovery rate (trough) __ ¹x.com/AnthPB/status/… ²AI-related fixed income issuance across IG, leveraged loans, and securitized – excluding e.g. $googl equity fundraising ³capitalizing leases = $675B operating + $175B finance + $82B not yet commenced (excluding multiyear purchase commitments ⁴including $amzn $googl $meta $msft (ex $orcl) ⁵"time-to-power" fixes that fast-track normal 5–7 year grid hookup – e.g. on-site gas turbines (15~20 GW), fuel cells, nuclear colocation, Bitcoin-mining conversions, and other behind-the-meter deals

Anthony Bardaro @AnthPB

10 months ago

/5 …TLDR: 1️⃣ AI "center of gravity" R&D/startup funding runway is effectively secured through 2026h2⁸ 2️⃣ AI "center of mass" datacenter capex is unfunded atm: i. The "Power Wall" energy crunch is already here⁹ ii. The "Funding Gap" cash crunch escalates

1 0 1 909 2

…important notes from Dylan and Dwarkesh pod¹ that a lotta tweeps miss – the cost-revenue arbitrage/margin remains immense for AI ROI and quantifying Chinese AI competitive disadvantages: 1️⃣ 1GW = $6B annual capex (upstream chips and tooling) = $28B compute lease² (midstream from HCPs to AI labs) = $100B downstream end user revenue (economic value created) 2️⃣ China vs US share of global compute (GW)³: •2022 (new flow): 🇨🇳30~35% vs 🇺🇸45~50% •2026 (new flow): 🇨🇳<10% vs 🇺🇸70%⁴ ⁵ •2028 (total stock/capacity): 🇨🇳<30GW vs 🇺🇸150GWe __ ¹youtu.be/aV26V1UvkJw?t=… ²part of this ~80% gross margin between that capex cost and lease price is part of why the cost competitiveness of open vs closed models is overstated – good luck bringing your own model to your own tiny MSA/colo/SaaS/API/on prem vs proprietary lab/HCP scale economies #BYOM #OSS #FOSS #SLA ³youtu.be/aV26V1UvkJw?t=… ⁴effects of US's China Chip Ban, which is working (x.com/AnthPB/status/…) and needs adaptation to plug holes (x.com/AnthPB/status/…) ⁵fodder for Jensen's Paradox (x.com/AnthPB/status/…)

Anthony Bardaro @AnthPB

5 months ago

@GavinSBaker @dwarkesh_sp /2 …but how does $nvda Jensen¹ resolve his own logical inconsistency between: 🅰️ China having effectively infinite "free" energy is their AI equalizer that offsets US AI chip superiority (even with trailing edge chips due to China Chip Ban) --VS-- 🅱️ The US selling more

1 0 2 1K 0

…re datacenter construction delays from supply chain shortages and bottlenecks: 🛠️ 30–50% of 2026CY datacenter pipeline unlikely to come online before 2026YE (vs 36%¹ in 2025CY)... ↳ 16GW announced for 2026CYe (140 projects total) 🟰 5GW currently under construction ➕ 11GW announced but not started ⚡️ "shortage of electrical equipment, such as transformers, switchgear and batteries… not enough domestic capacity to go around [so imports]… Electrical infrastructure adds up <10% of the total cost of the datacenter, but it’s impossible to build the operation without it"² __ ¹"in 2025, 26% of expected capacity slipped, and another 10% of projects pushed back their commercial operation dates without much notice" (sightlineclimate.com/research/data-…) ²bloomberg.com/news/features/…

keyboard_arrow_left Previous keyboard_arrow_right Next

Individualism and Momentum around the World (Chui/Titman/Wei, 2009¹): ‘Why does momentum work in some markets but not in others? more individualism > more uninformed trading > higher momentum returns’ __ ¹papers.ssrn.com/sol3/papers.cf… $study

Green Lark @the_green_lark

2 days ago

Why does momentum work in some markets but not in others? more individualism > more uninformed trading > higher momentum returns Chui, Titman, Wei "Individualism and momentum around the world"

9 6 115 16K 116

@borrowed_ideas icymi 👇 medium.com/adventures-in-…

Anthony Bardaro @AnthPB

3 weeks ago

🏁/15 ...TLDR (medium.com/adventures-in-…): • The Moderate Middle for AI (instead of doomers and boomers and the Horseshoe Theory union thereof) • Bardaro's Cost Disease⁶ (AI disinflationary, but inflationary for goods/services untouched by AI) • The Hippocratic Loathes

keyboard_arrow_left Previous keyboard_arrow_right Next

0 1 4 4K 1

...related: x.com/AnthPB/status/…

Anthony Bardaro @AnthPB

4 days ago

…echoing big August mtd recovery in high-frequency consumer card spending data (📸), shipping traffic through Strait of Hormuz has recovered notably (Goldman¹), which increases Trump leverage² in Iran War negotiations: 🚢 OIL THRU HORMUZ: 8~10 mbd 40~70% of prewar

keyboard_arrow_left Previous keyboard_arrow_right Next

1 0 1 173 1

…Bessent's Iran War TLDR – "economic D-Day…one-two punch [to] collapse this regime" means, effective tomorrow: 1️⃣ increased economic sanctions on Iran and secondary sanctions on anyone with Tehran ties (e.g. Iranian trade partners, banking, shipping, oil, etc) 2️⃣ preexisting naval blockade __ #EconomicFury #navy

Treasury Secretary Scott Bessent @SecScottBessent

a week ago

President Trump has dismantled Iran’s military capabilities, destroyed nearly 100 percent of its military factories, and buried its nuclear program. We are now entering the endgame. At dawn begins an economic D-Day — the single greatest financial offensive ever marshaled against

7K 10K 58K 5.5M 3K

"UN Security Council permanent member veto power..." 🙃 as the United Nations Security Council hosts an open session on Russian conflict in Ukraine... x.com/i/events/14959… #putin

...related: x.com/AnthPB/status/…

Anthony Bardaro @AnthPB

a week ago

…Bessent's Iran War TLDR – "economic D-Day…one-two punch [to] collapse this regime" means, effective tomorrow: 1️⃣ increased economic sanctions on Iran and secondary sanctions on anyone with Tehran ties (e.g. Iranian trade partners, banking, shipping, oil, etc) 2️⃣ preexisting

1 0 1 257 0

…echoing big August mtd recovery in high-frequency consumer card spending data (📸), shipping traffic through Strait of Hormuz has recovered notably (Goldman¹), which increases Trump leverage² in Iran War negotiations: 🚢 OIL THRU HORMUZ: 8~10 mbd 40~70% of prewar levels (close to US officials' estimates) 🛢️ OIL FROM BROAD PERSIAN GULF: 15~16 mbd ~66% recovery 🛍️ OTHER GOODS: weak (containers, dry bulk, chemicals, LNG, fertilizer, etc) __ ¹finance.yahoo.com/energy/article… ²axios.com/2026/08/28/tru… $USO $cl_f #MOU

keyboard_arrow_left Previous keyboard_arrow_right Next

…re "future tripwire" beyond which AI largesse could overwhelm and crowd-out real economy – here's THE RULE OF 25% fwiw¹: > "Over America’s 250-year history, the economy has absorbed [cumulative] spending on transformation technology equivalent to ~25% of [GDP] before things get really hairy...With US GDP at ~$30T, the danger zone sits at ~$7.5T, or an additional +$5T to +$6T in [cumulative] domestic AI spending. Hyperscalers are projected to spend [cumulative] +$3.7T globally through 2029 [excluding] Oracle, SpaceX, Anthropic, and OpenAI, among others [so] at this rate, AI spending won’t trip the rule of 25 until the early 2030s, six or seven years into the boom"² ³ __ ¹obvi not an economic law; comping cumulative capex to annual GDP is moving target; ignores historical context of duration, depreciation, import content, etc; but echoes my point that "slower physical rollout" governed by "natural breaks" would be a virtue ²x.com/trevornoren/st… ³barrons.com/articles/ai-ca… #capital #boom #bust

Trevor Noren @trevornoren

5 days ago

"Over America’s 250-year history, the economy has absorbed spending on transformation technology equivalent to about 25% of economic output before things get really hairy...With U.S. GDP at roughly $30 trillion, the danger zone sits at about $7.5 trillion, or an additional $5

5 16 67 6K 50

/3 …easy for anyone to point at empirical data² rn and say ‘looks like we're digesting AI well enough’ – but is there some current rot³ or future tripwire⁴ you have flagged? __ ²e.g. real consumer spending still up; AI capex carrying a large share of GDP growth; unemployment not spiking; equity wealth concentrated so the wage channel still matters ³e.g. maybe macro wage/consumption internals deteriorating vs headlines? ⁴e.g. maybe some number/ratio at which AI hits its sustainable growth rate, whether capex or funding stock/flow, where its scale crowds-out the rest of the economy or capital markets?

Philippe's question was ‘who consumes final goods when AI cuts payrolls’ and Dwarkesh answered ‘who are the AI users/what are the AI use cases’ – two important but very different points… and Barnett’s CES exercise has no macroeconomic closure (e.g. raising the supply of remote labor does not show that demand exists to absorb the extra output)… 1️⃣ I illustrated one such hypothetical closure (x.com/AnthPB/status/…) – labor-cost cuts can show up as lower prices, exports, profits/wealth, and financed capex, while bottlenecks slow substitution 2️⃣ I proposed policy solutions (x.com/AnthPB/status/…) – including package to address remote work and other effects

Anthony Bardaro @AnthPB

6 days ago

@IrvingSwisher /2 …e.g. an equal-but-moderate hypothetical "very specific/narrow kind of [good] outcome": labor-cost cuts need not cut nominal spending if they arrive as lower prices/export demand 👇 higher profits/wealth 👇 capital inflows refinance capex boom and wealth effect, while

1 1 2 519 1

…more on the returns to higher education – median value of various levels of educational attainment (income vs net tuition and opportunity costs)¹: HIGH SCHOOL: $1.65M median lifetime earnings ASSOCIATE (2-YEAR): +$145k NPV (over HS) +15.5% IRR (real, over HS) BACHELORS (4-YEAR): +$410k NPV (over HS) +13.3% IRR (real, over HS) MASTERS (+2 YEARS): +$615k " " " +9.5% " " " " DOCTORAL PHD (+5): +$520k +7.8% PROFESSIONAL MD/JD/DDS: +$980k +11.8% __ ¹share.gemini.google/2PJOGnZhSW6H #school #degree #diploma

/5 …afaik, rising admin support for public school kids attributable to: 1️⃣ No Child Left Behind Act and other federal/state regulations (reporting, testing, etc) 2️⃣ Specialized services (after school programs, expanded food/nutrition) 3️⃣ Rising IT needs 4️⃣ Safety/security

🧵/0 Q: Are kids learning? How do we measure effectiveness and outcomes of education in the US? #academics #elementary #schools #university

Closely Monitoring Lucky Lemur @TakNevimNevim

2 years ago

@AnthPB @GallupWorkplace @pewresearch @nytimes @EducationNext But is our children learning? How do we measure the outcome/usefulness of K-12 edu?

0 0 0 991 0

/2 …e.g. an equal-but-moderate hypothetical "very specific/narrow kind of [good] outcome": labor-cost cuts need not cut nominal spending if they arrive as lower prices/export demand 👇 higher profits/wealth 👇 capital inflows refinance capex boom and wealth effect, while slower physical rollout than Citrini's 2028 D-Day (due to natural breaks/bottlenecks/industrial capacity) 👇 households keep consuming via real income and wealth effects, while firms fund AI from savings and capital markets

/1 Yeah, that bear case kinda rhymes with Citrini's 2028 scenario, which would be a very specific/narrow kind of bad outcome, but there are natural breaks¹ to make that particular wormhole improbable, since temporally elongating AI/AGI/ASI adoption and diffusion to avert overnight shocks helps economic adaptation – so depending on what kinda future you're envisioning (other variables around "hypergrowth" constant), you're asking about how continued exponential AI hyperacceleration could theoretically sustain itself from both macro and micro perspective: 1️⃣ external sector and US absolute advantage > comparative advantage (via Heckscher-Ohlin Theory of Factor Abundance) 2️⃣ Bardaro's Cost Disease (x.com/AnthPB/status/… and medium.com/adventures-in-…) 3️⃣ Rockefeller and The Services Reallocation Flywheel, with Goldilocks creative destruction (x.com/AnthPB/status/…) 4️⃣ micro and macro ‘where does the money come from?’ (x.com/AnthPB/status/…) __ ¹x.com/AnthPB/status/…

Anthony Bardaro @AnthPB

7 months ago

/1 Base case expectation from AI's macro impact is a derivative of Baumol's cost disease (more like Balassa-Samuelson effect with echos of post pandemic bullwhip effect and ofc Jevon's Paradox): disinflation and rising per capita profit pools in AI-affected businesses produce

1 3 2 9K 2

…should be explicitly noted that Pax Silica¹ is the Trump Administration's full roundtrip from its reonshoring tactics/strategy² ³ (unilateral tariffs/go-it-alone domestic manufacturing and supply chains) back to Biden Admin trajectory⁴ (combo of reonshoring/nearshoring/friendshoring), which itself was in many way a continuation of Trump 1.0 policy except multilateral – regardless, all an adaptation and evolution that acknowledges the need to build alternatives/redundancies⁵ to China __ ¹en.wikipedia.org/wiki/Pax_Silica ²x.com/AnthPB/status/… ³x.com/AnthPB/status/… ⁴x.com/AnthPB/status/… ⁵x.com/AnthPB/status/… #trade

Anthony Bardaro @AnthPB

4 years ago

...more on chips "tradeoff between the inefficiency of that redundancy and the security of the excess capacity" – fm Mark Liu $tsm: "100% in one place cannot suffice...we started the overall global footprint planning...of course, the cost will be higher" stratechery.com/2023/tsmc-earn…

1 0 1 369 0

...more: x.com/AnthPB/status/…

Anthony Bardaro @AnthPB

a year ago

..."a few of the more prominent problems with Trump’s Liberation Day tariffs and this entire [diplomatic?] approach" to multilateral negotiations (forum.stratechery.com/t/trade-tariff…): 1️⃣ automation vs jobs and comparative advantage 2️⃣ duration mismatch 3️⃣ Trump equilibrium/disequilibrium

keyboard_arrow_left Previous keyboard_arrow_right Next

1 0 1 713 0

Given political polarization and ideology nowadays, it's actually heartening to see that anti-Chinese sentiment among Americans hasn't degraded to rival prior bouts of xenophobia/nationalism¹ ² – an outlier that US popular opinion has largely been able to distinguish between the CCP state³ and Chinese people⁴⁵⁶⁷ in the age of broad strokes and no nuance… __ ¹ofc many exceptions like COVID-era anti-Asian incidents and lingering "perpetual foreigner" biases, but relatively small share compared to history ²e.g. Chinese Exclusion era, Yellow Peril agitation, WWII treatment of Japanese Americans, etc ³pewresearch.org/short-reads/20… ⁴pewresearch.org/global/2008/06… (2008) ⁵researchgate.net/publication/22… (2010) ⁶pewresearch.org/global/2021/06… (2021) ⁷pewresearch.org/global/2022/09… (2022)

keyboard_arrow_left Previous keyboard_arrow_right Next

…to wit – despite "China having effectively infinite free energy [as] their AI equalizer that offsets US AI chip superiority"¹ and CCP nearshoring/friendshoring datacenter access², the US's China Chip Ban is taking a bite out of Chinese compute capacity – and offshore datacenter disintermediation matters wrt geopolitical calculus too: __ ¹x.com/AnthPB/status/… ²x.com/AnthPB/status/…

Anthony Bardaro @AnthPB

a week ago

…important notes from Dylan and Dwarkesh pod¹ that a lotta tweeps miss – the cost-revenue arbitrage/margin remains immense for AI ROI and quantifying Chinese AI competitive disadvantages: 1️⃣ 1GW = $6B annual capex (upstream chips and tooling) = $28B compute lease²

1 0 1 416 1

/5 …idk, trying to turn the tide on Sino-American diplomatic relations comes from a good place – and we're not exactly in the gilded age of logical consistency – but it's funny that Jensen's argument boils down to "reversing strategic decoupling", given Trump/MAGA/populist deglobalization is full speed ahead in the other direction 🤷