A Global South Diversification Strategy

I believe Singapore should reject binary alignment with either Washington or Beijing, not as a temporary hedge, but as a permanent structural doctrine, and I believe this doctrine should be backed by concrete diversification across critical minerals, reserve composition, institutional positioning, and formal trade targets across the Global South.

Part One: Trading American Dependence for Chinese Dependence Solves Nothing

Outright alignment with China trades dependence on one great power’s currency, capital markets, and political stability for dependence on another’s, with the same category of risk simply relocated.  China’s economy carries documented vulnerabilities: a property sector still working through Evergrande’s January 2024 Hong Kong liquidation and the broader sector’s US$300 billion in defaulted debt, a domestic consumption base structurally weaker than its export capacity, and a demonstrated capacity for abrupt, centrally imposed policy reversal, visible in the sudden December 2022 abandonment of zero-Covid following nationwide protests, and in the 20 per cent offshore trust tax Beijing imposed with a 90-day compliance window in July 2026.  I believe jumping from American overreliance to Chinese overreliance is not diversification.  It is the identical structural mistake wearing a different flag.

Part Two: The Non-Alignment Doctrine Already Exists

I did not invent this position.  Foreign Minister Vivian Balakrishnan has stated Singapore “does not take sides” but rather “upholds principles,” and has described our approach as “omni-directional engagement ... with all the multiple poles of power that are emerging.”  Speaking to Parliament in February 2026, he went further, stating Singapore would not act as “a proxy for any major power,” and must be prepared to “courteously stand up and say no,” a principle he applied to China as much as the United States.  Then Prime Minister Lee Hsien Loong stated on 1^st April 2022 that Singapore is not a US ally, will not conduct military operations on its behalf, and will not seek direct US military support, a position Prime Minister Wong Shyun Tsai reiterated in 2024.  When Donald John Trump imposed a 10 per cent baseline tariff on Singapore in April 2025, despite Singapore running a trade deficit with the United States, Wong stated Singapore was “very disappointed” and that “these are not actions one does to a friend.”

Foreign Policy magazine identified the risk: “The danger for the United States is not that Singapore suddenly pivots toward China.  It is that Singapore gradually diversifies its diplomatic, economic, and strategic relationships in ways that reduce US” leverage.  I believe this diversification is already under way, and I believe we should stop framing it, in official communication, as a reactive hedge adopted each time external pressure forces a response, and start stating it as the permanent structural doctrine it already is in substance.

Part Three: Singapore Should Not Pursue Reserve Currency Status

The Singapore dollar held its value through the 2019 Gulf tensions with only a 0.4 per cent decline, against 1.5 to 2.5 per cent losses among regional peers, and repeated this through 2026’s Iran-linked turbulence.  This makes Singapore Government Securities look, on paper, like a candidate for central banks diversifying away from concentrated Treasury holdings, especially given the dollar’s reserve share has fallen from over 70 per cent in 2000 to under 59 per cent by 2024, and central banks bought a record 288.9 tonnes of gold in a single quarter of 2026.

I do not believe Singapore should pursue this, and MAS’s decades-long policy already reflects why.  MAS has maintained a long-standing policy of actively discouraging internationalisation of the Singapore dollar.  A 1996 IMF working paper explained the reasoning: “The MAS frowns on internationalising the local currency because it believes that a large pool of Singapore dollars in the hands of non-residents can be a source of exchange rate instability.”  MAS manages inflation through the exchange rate, not the interest rate, the “only macroeconomic instrument” it uses to stabilise domestic prices.  A pool of foreign reserve demand for Singapore dollars would introduce persistent currency appreciation pressure unrelated to our domestic needs, forcing an impossible choice between an overvalued currency and continuous, self-defeating intervention.

Economist Robert Triffin’s 1960 dilemma applies here even in Singapore: supplying enough currency to satisfy global reserve demand typically requires running persistent external deficits, undermining the currency’s stability.  Singapore has built its AAA rating on the opposite discipline, a net asset position sustained by current account surpluses.  I believe this recommendation should be rejected, and MAS’s existing non-internationalisation stance affirmed as correct.

Part Four: Institutional Continuity is a Measurable Comparative Advantage

I believe Singapore’s institutional continuity is not a marketing claim, but a measurable, decades-long track record we have not yet stated with sufficient force.  The Economic Development Board was established in 1961.  The Housing and Development Board followed in 1960, now housing over 80 per cent of citizens under one continuous mandate.  GIC was established in 1981 and has run a single investment mandate for over four decades.  The World Bank’s Worldwide Governance Indicators score Singapore’s Political Stability index at an average of 1.27 points across 1996 to 2023, against a global average of negative 0.06, ranging narrowly between 0.88 and 1.6 across nearly three decades.  A range this narrow, over this long, shows not merely high governance quality but near-zero volatility in it.

This is already converting into measurable capital.  Singapore’s single family office count grew from 400 in 2020 to over 2,000 by 2026, assets under management approaching S$7 trillion, driven by capital fleeing China’s offshore trust tax, the UK’s non-domicile exodus, and Dubai’s conflict exposure.  Each source market shares the same driver: a sudden, discontinuous policy shift our institutional architecture does not produce.  I believe we should state policy continuity itself as the headline argument in economic diplomacy and EDB investment promotion, rather than leaving it implicit inside a list of Singapore’s more frequently cited attributes.

Part Five: The Concrete Critical Minerals Opportunity

Thirty years of reversed American policy on tax, infrastructure, and industrial strategy stand in contrast to China’s decade-long planning cycles, visible in its consolidation of rare earth markets.  China controls 90 per cent of global rare earth processing capacity, over 60 per cent of lithium refining, and imposed export licensing on seven heavy rare earth elements in 2025, pushing European prices to six times Chinese domestic levels.

I see a concrete opportunity here, built on treaty architecture we already hold.  Singapore and Australia signed the Singapore-Australia Green Economy Agreement and Digital Economy Agreement, both in force.  Both countries are part of RCEP and CPTPP.  Australia holds substantial lithium reserves but limited processing capacity.  Indonesia controls the world’s largest nickel reserves, though 75 per cent of its refining capacity is already controlled by Chinese firms following over US$30 billion in Belt and Road-linked investment.  Australia’s counterbalancing effort, Nickel Industries’ US$1.7 billion investment at Morowali Industrial Park, remains a fraction of Chinese capital in the same sector.  I believe a Singapore-based trading and financing intermediary, built under our existing treaty framework, could aggregate Australian lithium output, blend it with non-Chinese-controlled Indonesian nickel supply, and route it toward AI infrastructure buyers across CPTPP markets, without a single new treaty.  The window is real because Western reshoring remains structurally years away: the EU’s Critical Raw Materials Act targets only 10 per cent domestic extraction by 2030, and Washington’s US$12 billion Project Vault exists to stockpile minerals, not build processing capacity.

Part Six: A Formal, Measurable Global South Target

Singapore already applies precise, tracked metrics to bilateral trade.  Trade in goods with Latin America more than doubled over five years to over S$35 billion in 2025.  The Pacific Alliance-Singapore Free Trade Agreement entered into force in May 2025.  Our bilateral trade in goods with the UAE reached S$24 billion in 2024, with services trade growing 22 per cent year-on-year, and Singapore’s investment into the UAE grew 13 per cent to S$4.9 billion in 2023.  We have maintained a Gulf Cooperation Council free trade agreement since 2013.

The gap is with Africa.  Africa’s trade with Southeast Asia amounted to just 2.2 per cent of the region’s total world trade in 2021.  African investment into ASEAN totalled just US$188 million that year, 0.1 per cent of ASEAN’s total inward investment.  Then Minister Gan Kim Yong himself acknowledged this at the Africa Singapore Business Forum, stating investment flows “pale in comparison to the opportunities available.”  I believe MTI and MAS should answer that question by formalising a single, published Global South diversification target, tracked with the same specificity already applied to individual bilateral FTA performance, and reported on a fixed schedule Parliament can hold the government to.

My Conclusion

I have made the case that Singapore should reject binary alignment permanently and explicitly, not pursue reserve currency status despite its surface appeal, state our institutional continuity as the headline diplomatic asset it already is, build concrete critical minerals infrastructure on treaties we already hold, and set a formal, measurable Global South diversification target where the gap, particularly with Africa, is already documented and already raised in our Parliament.  None of this requires choosing a side in the cycle described in my first paper.  I believe it requires building the infrastructure that lets Singapore keep functioning regardless of how that cycle resolves.

---

Terence Nunis | Executive Chairman, Equinox Zenith | Author, The 1% Playbook: The Billionaire Cheat Code

The Case for an Independent Blue Carbon Exchange, Built to Fund the Loss and Damage Fund

I believe the Loss and Damage Fund, established to help developing countries adapt to climate change, has failed on its terms, and I believe an independent, investment-grade compliance carbon exchange is the most credible instrument available to fund it properly.  Here, I make that case, and contend that the Loss and Damage Fund is the explicit strategic endpoint of everything that follows, not an implicit byproduct of building better market infrastructure.

Part One: The Loss and Damage Fund Has Failed

The Loss and Damage Fund closed COP28 with pledges just over US$600 million, a sum that was smaller than the cost of building the Dubai Expo City venue hosting the conference.  By September 2024, pledges reached US$702 million from 23 contributors.  The UN itself estimates actual need at US$300 billion a year by 2030, rising to US$500 billion by 2050.  The gap between pledge and requirement is 400 to 1.

The pattern has only worsened since.  The United States rescinded US$4 billion in Green Climate Fund pledges in February 2025, the first country ever to formally withdraw a commitment already made.  The United Kingdom halved its pledge in spring 2026.  A planned pledging event at COP30 for the Least Developed Countries Fund and Special Climate Change Fund was cancelled outright in November 2025, for lack of contributor interest.  The World Bank dropped its 45 per cent climate co-benefits target in June 2026, the very month it had already exceeded that target at 48 per cent, under pressure from the United States, Russia, and Saudi Arabia.

I believe public, pledge-based climate finance is structurally unreliable, because it depends on donor governments whose domestic politics can reverse a commitment at any point, with no penalty for doing so.

Part Two: A Secondary Market Solves the Reliability Problem a Pledge Cannot

Article 6’s rulebook, finalised at COP29 in November 2024, and the Paris Agreement Crediting Mechanism, operational following COP30 in November 2025, give carbon credits legal infrastructure to trade as a transferable financial asset rather than a voluntary gesture.  The EU Emissions Trading System has already proven this model at scale, cutting covered emissions 51 per cent since 2005 while raising over €265 billion in revenue, funded entirely by market participants pricing their carbon output, with no pledging conference and no government able to walk the commitment back.

I believe a market-priced instrument, once built properly, does not depend on the next election cycle in Washington, London, or anywhere else.  This is the structural advantage a secondary carbon market holds over every pledge-based mechanism, and it is the reason I am proposing this exchange as a Loss and Damage funding vehicle, not as a Singapore commercial opportunity.

Part Three: The Verification Process is Broken

Verra, the dominant voluntary registry, lets project developers hire and pay their auditors directly.  Science magazine’s editorial board stated, “Auditors are unlikely to stay in business if they disapprove credits at the high rates that research suggests would be appropriate today.”  Carbon Market Watch found 21 of 33 accredited auditors active in 2024 had signed off on at least one of 95 projects later found to have overstated their climate benefit.  Transparency International US described the resulting arrangement as “students designing their assignments and grading their papers.”

PACM’s first issuance, a Myanmar cookstove project approved in February 2026, was subsequently found to have been approved for roughly seven times more credits than its actual emission reductions likely warranted.  It operated through institutions controlled by Myanmar’s military junta, in conflict-affected regions, with verifiers unable even to conduct site visits due to security concerns.  SK Telecom, the buyer relying on these credits, has itself publicly acknowledged the claimed reductions cannot be verified.

I believe this is no longer a niche technical criticism.  It is a documented, publicised failure spanning both the dominant voluntary registry and the UN’s flagship compliance mechanism, and it gives institutional buyers a concrete, citable reason to seek a genuinely independent alternative.

Part Four: The Structural Fix

Whoever verifies a credit’s emission reduction claim must have no financial relationship with the project developer seeking that credit approved.  A verification architecture paid by the exchange, by an independent buyer-side consortium, or through a standing endowment rather than per-project developer fees removes the incentive that is the root cause of the Verra pattern.

Mangrove and seagrass carbon sequestration suits this model in a way human, developer-paid auditors structurally cannot match.  Satellite-based remote sensing can measure canopy coverage, biomass density, and sequestration rate directly, independent of any relationship between verifier and developer, and independent of the security access constraints that left Myanmar’s verifiers unable to confirm anything in person.  I believe applying AI-based MRV to blue carbon specifically converts a verification method that failed catastrophically in cookstove methodology into one resistant to the same failure mode, because the underlying claim is measurable from orbit rather than dependent on a developer’s self-reported survey data.

Part Five: The Legislation and Compliance Architecture Required

Internationally, Article 6.2’s corresponding adjustment mechanism prevents double counting, requiring the host government to formally record the transfer in its national emissions inventory.  A Letter of Authorisation must be issued by the host state and verified by the relevant carbon standard before any credit can trade internationally.  CORSIA’s compliance phases already require this authorisation as a precondition for airline offset eligibility.  AI-based MRV for blue carbon would need to satisfy the Article 6.4 Supervisory Body’s methodological standards, currently scoped narrowly to methane flaring and nitrous oxide abatement, meaning a formal new methodology submission and approval process precedes any blue carbon-specific AI verification protocol gaining recognition under the compliance-grade international framework.

Domestically, I believe Singapore’s regulatory architecture already provides usable scaffolding.  The Variable Capital Companies Act 2018 allows a fungible carbon credit exchange to be structured with segregated sub-funds, ring-fencing different project pools, vintages, or geographic sources, administered by the Accounting and Corporate Regulatory Authority with anti-money laundering oversight from MAS.  A verification layer built independently of the credit issuer would need to satisfy MAS’s capital markets licensing framework under the Securities and Futures Act if it intends to offer credit-linked derivatives or structured products.  Any blue carbon project sourcing credits from Southeast Asian coastal states would need bilateral Article 6 authorisation agreements negotiated with each host government individually, since PACM and Article 6.2 operate project by project and country by country, not through a single blanket regional approval.

Part Six: Singapore’s Existing Position

As of June 2026, Singapore has signed Article 6 Implementation Agreements with eleven countries: Papua New Guinea, Ghana, Bhutan, Chile, Peru, Rwanda, Paraguay, Thailand, Vietnam, Mongolia, and the Philippines.  In September 2025, Singapore contracted 2.175 million tonnes of nature-based carbon credits from projects in Ghana, Peru, and Paraguay.  On 6^th July 2026, Singapore signed a major bilateral agreement on carbon credit collaboration with Indonesia, witnessed by Indonesian President Prabowo Subianto and Singapore Prime Minister Lawrence Wong Shyun Tsai, establishing a corresponding adjustment framework for cross-border trading.

Singapore’s commissioned study by the Economic Development Board and Enterprise Singapore estimates the carbon services and trading hub ambition could generate US$1.8 to US$5.6 billion in gross value added.  That study assumed Singapore’s continued positioning as the region’s default hub, an assumption now contestable as Indonesia, Vietnam, and Malaysia each build their domestic registries and exchanges.  I believe the task ahead is not building new bilateral relationships from scratch, largely complete, but building the interoperability layer that keeps Singapore central as these national systems mature, through active participation in the ASEAN Common Carbon Framework the ASEAN Alliance on Carbon Markets is already developing.

My Conclusion

A market-priced, independently verified compliance carbon exchange, built specifically for blue carbon and verified through AI-based remote sensing rather than developer-paid human auditors, is not merely a better carbon market.  It is, I believe, the most credible mechanism currently available to close the gap between the US$300 billion a year developing countries need and the US$700 million the pledge-based Loss and Damage Fund has managed to raise.  Singapore already holds the treaty architecture, the regulatory scaffolding, and a documented lead in bilateral Article 6 agreements to build this.  What it does not yet have is the independent verification layer, and I believe that is the single piece of infrastructure Singapore should now be directed toward building.

---

Terence Nunis | Executive Chairman, Equinox Zenith & Red Sycamore | Author, The 1% Playbook: The Billionaire Cheat Code

Reassessing Singapore’s Sovereign Exposure to the American AI Bubble

I believe a debt-financed capital loop is inflating an American AI bubble to a scale the underlying cash flow does not support.  I believe that bubble sits inside an economy that no longer carries the fiscal reserves it once had to absorb a correction.  I believe the same debt trajectory is pushing Treasury yields to levels the market itself is beginning to call structural rather than cyclical.  I believe China’s chip constraints, intended to slow its AI progress, instead forced a more efficient architecture that may prove the more durable advantage.  I am making the case that Singapore’s sovereign funds should reassess their exposure accordingly.

A circular financing loop, estimated at over US$800 billion, links Nvidia, AI laboratories, and cloud providers in a structure that books related-party spending as independent revenue.  OpenAI alone has committed US$1.15 trillion through 2035 against a projected US$14 billion loss in 2026.  Nearly 40 per cent of the S&P 500’s market capitalisation is now under direct AI influence.  Lucent, Nortel, 1929, and 2008 each followed the identical pattern: concentrated leverage mistaken for collateral, collapsing once independent cash flow failed to materialise.  Federal debt has crossed US$40 trillion, and the Federal Reserve raised rates in September 2026 rather than cutting them, removing the monetary cushion that absorbed the 2001 dot-com correction.  The 10-year Treasury yield crossed 5 per cent the same month, and every Treasury intervention attempted in 2026 produced only temporary relief before yields resumed climbing.  Chip sanctions forced Chinese laboratories toward architectures requiring a fraction of the compute of Western models, at a fraction of the cost.  US data centres already consume as much electricity as Ireland’s entire national grid, with household electricity bills already rising to fund the buildout.  The top 1 per cent of US households now hold 31.7 per cent of national wealth, concealing fragility beneath resilient-looking aggregate consumer data.  Each section below closes with a specific implication for Singapore’s funds.

Part One: The Capital Loop Feeding an Unsustainable Bubble

Nvidia invests billions into AI laboratories such as OpenAI and Anthropic.  Those laboratories sign compute contracts with cloud providers, including Microsoft, Oracle, and Amazon Web Services.  Those providers spend a large share of that revenue buying chips back from Nvidia.  Cash leaves Nvidia’s balance sheet as an investment.  It returns as revenue, having toured through two or three other balance sheets along the way.  Analysts have identified over US$800 billion moving through this loop, and deals of this scale fuel circularity concerns.

OpenAI has committed US$1.15 trillion across seven vendors through 2035, while projecting a US$14 billion loss for 2026, nearly triple the prior year’s loss.  Big Tech tripled annual AI capital spending from US$150 billion in 2023 to over US$500 billion in 2026.  Nearly 40 per cent of the S&P 500’s market capitalisation is under direct AI influence, a concentration ratio without precedent outside the dot-com peak.  Forward price-to-earnings on the index is 22.85 times, within a fraction of the 2020 peak of 23.11.

Revenue booked from a related-party compute contract is not the same as revenue earned from an independent customer with an independent reason to keep paying.  When one node in this loop misses a covenant or writes down an asset, the loss does not stay contained to that balance sheet.  It removes demand from every counterparty whose revenue projections assumed the spending would continue uninterrupted.

Temasek Holdings reported a net portfolio value of S$518 billion, US$401 billion, for the year ended 31^st March 2026.  The United States accounts for 26 per cent of that portfolio.  AI-related exposure is 6 per cent and is committed to rise to 10 to 15 per cent by 2031.  Temasek Holdings holds stakes in both OpenAI and Anthropic, and was, per Reuters, in active discussions to increase its OpenAI position further.  Chief Executive of Temasek Global Investments Chia Song Hwee has told reporters overvaluation in AI is “unavoidable” and that nobody, Temasek Holdings included, can predict when a correction arrives.  Temasek Holdings has declined to disclose specific stakes or their effect on overall performance, leaving me unable to independently assess how concentrated this exposure has become beneath the reported 6 per cent figure.  I believe a portfolio review distinguishing genuine end-user revenue from circular vendor-financed revenue should precede any further allocation toward the stated target.

Part Two: The Historical Parallel, and the Write-Down Cascade

Lucent Technologies and Nortel Networks ran an almost identical loop during the dot-com era, lending customers money to buy their equipment and booking the proceeds as revenue on both sides.  When real demand failed to match financed demand, both collapsed in the same downturn.  Lucent’s market value fell from over US$258 billion at its 2000 peak to a fraction of that within two years.  Nortel filed for bankruptcy in 2009.

The 1929 crash carried the same architecture in a different sector: margin-financed speculation, concentrated in a narrow set of favoured stocks, collapsing once credit tightened.  The 2008 crisis ran the identical script through mortgage-backed securities.  Lehman Brothers held US$639 billion in assets at its collapse, and the resulting deleveraging wiped out an estimated US$19.2 trillion in US household wealth by 2011.  Every one of these episodes shares the same signature: concentrated leverage, mistaken for collateral, until the underlying cash flow failed to materialise.

MAS has built genuinely sophisticated AI governance infrastructure.  The MindForge AI Risk Management Toolkit, released in April 2026, was developed with 24 institutions, including DBS, OCBC, UOB, GIC, and Temasek, alongside Nvidia, AWS, Google Cloud, and Microsoft.  Director Alan Lim of MAS’s Financial Infrastructure and AI Office described it as moving institutions “from theory to practice.”  This framework governs operational AI risk, how banks deploy AI internally, not the separate question of what happens to Singapore’s banking balance sheet if the AI sector itself suffers a credit event.  GIC led Anthropic’s US$30 billion Series G round, valuing the company at roughly US$380 billion.  GIC took a leading role in financing a private AI company at a valuation built on projected revenue rather than demonstrated cash flow.  I believe MAS’s stress-testing framework for DBS, OCBC, and UOB should explicitly model an AI-sector credit event as a distinct scenario, separate from MindForge’s operational governance work.

Part Three: Why America Cannot Absorb This the Way It Absorbed the Dot-Com Crash

The dot-com correction in 2001 hit an economy with far more fiscal room.  Federal debt was below 60 per cent of GDP, and the government ran a surplus as recently as fiscal year 2000.  That room no longer exists.  National debt has crossed US$40 trillion.  The Congressional Budget Office projects debt held by the public rising from 101 per cent of GDP this year to 120 per cent by 2036.  Net interest reached US$963 billion over ten months of fiscal 2026, US$3.18 billion a day.

The Committee for a Responsible Federal Budget estimates that if yields stay just 80 basis points above baseline, annual interest payments reach US$2.7 trillion by decade’s end, exceeding combined Medicare and Social Security spending.  The Federal Reserve raised rates on 16^th September 2026, its first hike since 2023, because inflation had run above target for years and the central bank had no room to accommodate a fresh shock stacked on its own.

Singapore held US$280 billion in US Treasury securities as of April 2026, according to US Treasury International Capital data, separate from GIC’s estimated US$936 billion to US$1.16 trillion and Temasek Holdings’ S$518 billion portfolio, both carrying further undisclosed US exposure.  Singapore’s Ministry of Finance will not disclose GIC’s exact assets, because doing so would “amount to publishing the full size of Singapore’s financial reserves,” what it calls Singapore’s “war chest.”  Then MAS Managing Director Ravi Menon described the reserves’ three functions in 2019: a crisis buffer, an income stream, and a foundation for exchange-rate confidence.  A Treasury market seizing up around an AI-sector credit event does not merely threaten the market value of these holdings.  It threatens the function they exist to perform during a crisis.  GIC’s 20-year annualised real return fell to 3.4 per cent as of March 2026, a six-year low.  I believe MAS and MOF should explicitly model a US Treasury market dysfunction scenario tied to an AI-sector credit event, before market conditions force the question.

Part Four: The Structural Yield Problem

The 10-year Treasury yield crossed 5 per cent in September 2026.  The 30-year yield reached its highest level since 2001 the month before, with auction demand weaker than dealers expected.  The Treasury Borrowing Advisory Committee flagged a US$1.45 trillion funding shortfall for fiscal 2027 to 2028, a warning from the dealers who underwrite these auctions themselves.

Treasury Secretary Scott Kenneth Homer Bessent attempted three separate interventions across 2026: a joint yen operation with Japan structured to avoid selling Treasuries outright; a bond buyback funded by selling short-dated debt; and a described strategy of trading on “asymmetric information,” a hedge fund framing applied to sovereign debt stewardship.  Every intervention produced a temporary rally followed by a return to, or a breach of, the prior yield level within weeks, itself evidence that this is structural rather than cyclical.

I want to credit GIC before I criticise it.  Its 2025/26 annual report confirmed increased allocations to “commodities, gold, and infrastructure” specifically “to enhance inflation resilience,” and it overhauled its entire investment framework this year, replacing a 65/35 equities-bonds reference portfolio it had run for thirteen years with a Strategic Portfolio built around growth, uncertain rate navigation, and inflation resilience.  GIC’s stated reasoning names “a changing world order, rising fiscal risks, and advances in artificial intelligence,” the same three forces I have argued throughout this paper.

Here is the contradiction I believe needs resolving.  GIC’s equities allocation rose to 51 per cent in the year to March 2025, up from 46 per cent the year before, even as its report acknowledged high valuations “provide a challenging backdrop for forward returns.”  GIC has confirmed it “invested selectively” in AI, in companies it believes have “enduring value beyond short-term market enthusiasm,” an acknowledgement that it distinguishes durable value from speculative enthusiasm and has chosen to increase exposure regardless.  Global SWF, an independent tracker, has noted GIC’s reporting “has grown more opaque, now lacking specific asset class mix disclosures” that earlier reports provided.  I cannot verify what share of the 51 per cent equity allocation carries AI concentration, nor how the real assets increase compares in scale to the simultaneous equity increase.  I believe GIC and MAS should resolve this contradiction directly, and should, at minimum, internally restore the granular disclosure that Global SWF notes is no longer available.

Part Five: Why China’s Chip Constraints Handed It the Efficiency Advantage

US chip sanctions, intended to slow Chinese AI progress, instead forced a different kind of innovation.  DeepSeek released its R1 model in January 2025, matching OpenAI’s o1 on multiple benchmarks at a reported training cost under US$6 million, using export-controlled H800 chips.  Nvidia lost approximately US$589 billion in market capitalisation on 27^th January 2025, the largest single-day value loss in stock market history.

DeepSeek’s V3 model cost US$5.58 million to train, a 5.5 per cent fraction of GPT-4’s reported cost.  By 2026, this pattern spread across Alibaba’s Qwen, Moonshot AI’s Kimi, and Zhipu AI’s GLM.  GLM-5, released in February 2026, became the first frontier-scale model trained entirely on Huawei’s Ascend 910B chips, without a single Nvidia GPU.  DeepSeek founder Liang Wen Feng described the underlying constraint in a July 2024 interview: “We have to consume twice the computing power to achieve the same results ...  Our goal is to continuously close these gaps.”  His team closed them through architecture, sparse Mixture-of-Experts routing, FP8 low-precision training, aggressive context caching, cutting inference costs by 90 to 97 per cent against comparable Western models, not through acquiring more hardware.  In a world where US data centre electricity demand is projected to reach 325 to 580 terawatt-hours by 2028, an architecture requiring a tenth of the compute for equivalent output is the actual efficiency frontier, not a footnote to it.

Singapore’s National AI Strategy update, unveiled by Minister for Digital Development and Information Josephine Teo Li Min at ATxSummit on 20^th May 2026, names “efficient AI computing” as a stated priority.  Yet Singapore committed over S$1 billion in AI research funding, S$150 million to the Enterprise Compute Initiative, and S$37 billion under RIE2030, while OpenAI committed S$300 million to its first overseas Applied AI Lab here and Nvidia opened its second Asia-Pacific research lab here.  There is a stated commitment to efficiency, alongside deepening operational dependence on the same Nvidia supply chain already exposed to the rare earth export controls.  Singapore’s published strategy does not yet specify how efficiency translates into a concrete hedge against this dependency, as opposed to simply meaning lower costs within an unchanged hardware relationship.  I have also looked at AiRTS Pte.  Ltd., a Singapore-founded company building a patented enterprise product called GenAI Twin, as an example of genuine domestic AI intellectual property.  AiRTS operates at the application layer, addressing hallucination and output consistency in deployed models, not at the foundational training layer where DeepSeek’s efficiency gains are.  It is evidence of a capable domestic ecosystem.  It is not a substitute for the hardware diversification I believe the national compute strategy needs.

Part Six: The Domestic Precedent

Singapore imposed a de facto data centre moratorium from 2019 to 2022, after data centres reached 7 per cent of national electricity consumption, projected to climb to 12 per cent by 2030 without intervention.  Rather than simply lifting the ban, Singapore rebuilt the approval framework around efficiency as the primary gate.  The Data Centre Call for Application 2, launched in December 2025, requires every approved facility to achieve a Power Usage Effectiveness of 1.25 at full load, at least 50 per cent green power sourcing, and liquid cooling for at least 60 per cent of IT load, standards that significantly outpace regional benchmarks in Hong Kong, Japan, and South Korea.

This standard has not repelled capital.  Amazon Web Services pledged S$12 billion under this framework.  Google committed US$5 billion.  Jurong Island will host a 700-megawatt low-carbon data centre park.  Singtel’s Nxera unit is building a 58-megawatt facility backed by a US$476 million green loan.  Keppel plans to double its power capacity by 2030.

I believe any Singapore-linked data centre or compute cluster development outside our borders, in Johor, in Indonesia, or in any regional partnership our capital participates in, should be held to the same PUE and green energy standard we already enforce domestically.  A Singapore-backed facility built to a looser regional standard carries higher exposure to both rare earth cost inflation and water and power constraints, for a lower return on the same capital, than a facility built to our proven standard.  The domestic policy already exists.  I believe we should simply apply it consistently to where our capital builds next.

Part Seven: Unmanaged Data Centre Growth is a Quantified American Risk

PJM Interconnection’s capacity auction clearing price rose from US$28.92 per megawatt-day in the 2024/25 delivery year to US$329.17 for 2026/27, a factor of more than ten.  Data centres accounted for 63 per cent of that increase, translating to roughly US$9.3 billion passed to ratepayers.  New Jersey residents saw bills jump 17 to 20 per cent in a single year.  Virginia’s Dominion Energy implemented its first base-rate increase since 1992, adding US$8.51 a month to a typical household bill, in a state where data centres already account for 40 per cent of total electricity consumption.  Utilities requested US$18.6 billion in rate increases in the first half of 2026 alone, on pace to exceed 2025’s record US$29 billion.  Seven major technology companies signed a voluntary Ratepayer Protection Pledge in March 2026, committing to cover their grid costs.  The pledge carries no legal weight and cannot override existing tariff structures.

I believe Singapore should present this specific, quantified American contrast, not the general principle of sustainability, in every regional and international forum where data centre policy is discussed.  Our DC-CFA2 framework mandates efficiency before approval is granted.  America’s experience shows precisely what happens without that mandatory gate.

Part Eight: The Hollowed-Out Middle Class is a Financial Risk to Any AI-Exposure Model

The top 1 per cent of US households held 31.7 per cent of national wealth in the third quarter of 2025, the highest share since the Federal Reserve began tracking this in 1989.  That 1 per cent held US$55 trillion, close to the combined wealth of the bottom 90 per cent.  Wage growth for higher-income households reached 3 per cent in December 2025, against 1.1 per cent for lower-income households.  Moody’s Analytics found the top 10 per cent of households drove 49.2 to 49.7 per cent of consumer spending by mid-2025, though this figure has been disputed; the Bureau of Economic Analysis’s data show the top 10 per cent by disposable income responsible for only about 20 per cent of spending across 2004 to 2022.  I believe this dispute itself is evidence worth acting on: if credentialed economists working from the same Federal Reserve data cannot agree on how concentrated spending has become, the aggregate consumer spending figure is less reliable as a signal of broad economic health than most headline coverage suggests.  The wealth concentration data carries no comparable dispute, sourced directly from the Federal Reserve’s Distributional Financial Accounts: the top 1 per cent held 29.2 per cent of aggregate wealth in Q4 2025, against 5.3 per cent for the entire bottom half.

I believe any Singapore-based fund or institution modelling US consumer demand resilience should explicitly disaggregate spending data by wealth decile rather than relying on the aggregate figure, since that aggregate materially overstates broad-based economic health in an economy this concentrated, and since the same equity concentration directly transmits into this same consumer spending pattern through the wealth effect.

My Conclusion

I have traced one argument across eight parts.  A circular capital loop, an absent fiscal cushion, a structural yield problem, and a hollowed consumer base sit underneath an American AI bubble our sovereign funds are increasing exposure to, even as our central bank’s risk framework does not yet cover the specific scenario that exposure creates.  China’s constrained path produced a structurally more efficient alternative.  I believe Singapore’s institutions should reassess this exposure with the data, not with the assumption that scale alone protects against a correction nobody, including Temasek Holdings’ and GIC’s leadership, claims to be able to time.

---

Terence Nunis | Executive Chairman, Equinox Zenith | Author, The 1% Playbook: The Billionaire Cheat Code

Appointment Booster 3rd Anniversary: My Own Panel Answers on PIL (III)

I sat on the panel at AIA’s Appointment Booster 3^rd Anniversary on 9^th September 2026.  These are my own positions, expanded with anonymised examples drawn from my own client proposals.  The product in question was AIA Platinum Indexed Legacy (III), AIA Singapore’s universal life offering.

1.  What Made You Start?

The opportunity was never hidden.  It was sitting in plain sight; in every client conversation I was already having about tax exposure and cross-border wealth.  China’s Announcement No. 21 imposed a 20 per cent tax on offshore trusts from 24^th July 2026, with a 22^nd October deadline forcing families to declare or restructure.  The Gulf war unsettled Dubai as a haven.  Malaysia, Indonesia, and Thailand each carry their own political and currency uncertainty.  None of this is abstract.  It is what my clients were already telling me kept them awake.

One case in point: a senior technology executive, based in Singapore, held equity compensation, concentrated investment positions, and a philanthropic intent, but no structure connecting the three.  I built a multi-instrument stack around his position, combining an indexed universal life policy with investment-linked products and a self-funding philanthropic loop using Institution of a Public Character tax deductions under Section 37(3)(b).  PIL (III) was never a single product pitch.  It was an architecture problem, and once I started seeing client situations as architecture rather than product fit, the opportunity stopped being niche.

2.  Customer Objection Challenge: “I already have a lot of insurance.  Why do I need to look at this?”

I never open by arguing the client needs more insurance.  I open by asking what the insurance he already holds is for.  Protection covers loss of income.  Accumulation builds capital over time.  Legacy planning answers a different question entirely: how does the wealth already built get transferred, on what timeline, to whom, and under what tax exposure?

I built an objection-handling framework specifically for this moment: Acknowledge, reframe, validate.  Acknowledge the existing coverage, rather than dismissing it.  Reframe the conversation from “more insurance” to “does this achieve your stated legacy goal?”  Validate the gap once it surfaces, using the client’s own words rather than mine.  In one UHNWI proposal built around CRS 2.0 exposure, I used the AIG 2008 collapse as a case study, illustrating how concentrated institutional exposure, however well-insured on paper, can still leave a legacy plan structurally exposed if the underlying wrapper was never designed for cross-border succession in the first place.  Most clients holding a large book of protection products have never had that specific conversation.  I am not selling more insurance.  I am identifying whether what he already owns actually achieves what he says he wants for his family.

3.  Consultant Objection: “I don’t have HNW clients.  PIL (III) isn’t for my market.”

A legacy need does not require ultra-high-net-worth status.  It requires a client with any asset – a business, a property, a concentrated investment position – that he wants to pass on in a structured way rather than a single lump sum.  My own UHNWI materials, built around a Jumbo IUL structure, started life as a proposal for a single client.  I did not wait for a full book of accredited investors before building it.  I built the framework once, for one suitable case, and every subsequent proposal drew on that same glossary, objection framework, and case study library, refined rather than rebuilt from scratch each time.

4.  Practical Tip for Consultants: What to Prepare Before Your First PIL Presentation?

Understand the client before you understand the product further.  For the technology executive case, preparation meant mapping his equity compensation structure, his existing insurance stack, and his stated philanthropic intent, before a single slide on AIA Platinum Indexed Legacy (III) or AIA Pro Achiever 3.0 was ever shown.  The product mechanics matter, and a consultant should know them cold, but the opening conversation is never about the product.

It is about two or three precise questions: what happens to this asset when you are no longer here to manage it, who receives it, and on what timeline.  Prepare those questions, identify one suitable client to ask them, and do not wait until you feel completely ready.  Readiness is a feeling.  A scheduled conversation is a fact, and the fact moves the case forward considerably faster than the feeling ever will.

5.  Final Short Question: “If you haven’t started selling PIL (III) yet, just ______.”

Ask one client the legacy question.  Not the pitch.  The question.  Everything else, the framework, the case studies, the product stack, follows from there.

---

Terence Nunis | Executive Chairman, Equinox Zenith | Author, The 1% Playbook: The Billionaire Cheat Code

Quora Answer - What Do the New Fiscal Rules Announced During the UK Budget 2024 Mean for the Economy?

The following is my answer to a Quora question: “What do the new fiscal rules announced during the UK Budget 2024 mean for the country’s growth strategy?”

The new fiscal rules were introduced during the Autumn Budget 2024, presented to Parliament on 30^th October 2024, aiming to balance the current budget so day-to-day costs are met by revenue, with borrowing reserved for investment.  The rule targets public sector net financial liabilities, net financial debt as a share of the economy, meant to let public investment support growth while keeping debt under control.

The original framing held that stability would attract private investment and fund infrastructure, research, and development.  The market reaction at the time was mixed.  Equities rose.  The pound firmed.  Gilt yields rose too, reflecting concern over increased borrowing.

What Happened Since

Chancellor of the Exchequer Rachel Jane Reeves delivered a second budget on 26^th November 2025, raising taxes by £26 billion.  Growth forecasts were downgraded across every remaining year of the decade: 2026 cut from 1.9 to 1.4 per cent, 2027 from 1.8 to 1.5 per cent, 2028 from 1.7 to 1.5 per cent, 2029 from 1.8 to 1.5 per cent.  Debt itself, the metric these rules existed to control, is now projected to rise from 95 per cent of GDP to 96.1 per cent by the end of the decade, not fall.  The budget extended the freeze on personal tax thresholds to April 2031, a policy Reeves had explicitly promised would be a breach of Labour’s own manifesto if extended.  The Office for Budget Responsibility calculates the cumulative cost of that freeze, since its introduction in 2022-23, at £66.6 billion, the largest tax rise in sixty years.

The Spring Statement on 3^rd March 2026 cut the 2026 growth forecast again, from 1.4 to 1.1 per cent, with Reeves citing the escalating Middle East conflict as compounding an already uncertain outlook.  This is the second downgrade inside five months.

The Arithmetic That Matters

Debt interest costs reached £111.2 billion in 2025-26, 8.3 per cent of total public spending and 3.7 per cent of GDP.  The OBR’s own analysis notes the effective interest rate on UK government debt now exceeds the economy’s likely nominal growth rate, the condition under which debt-to-GDP tends to rise mechanically, regardless of new borrowing decisions.  Underlying public debt has grown 24 per cent of GDP over the past fifteen years, despite eight of the UK’s nine fiscal frameworks since 2010 explicitly targeting a falling debt ratio as their central goal.  The UK ran the fifth-highest budget deficit among 36 advanced economies in 2024.  Left on current policy, the OBR projects public debt could exceed 270 per cent of GDP by the early 2070s.

None of this unfolds in isolation.  The 2026 Sunday Times Rich List recorded Britain’s billionaire count falling to 157, with one in six families who appeared two years earlier gone, a documented wave of wealth relocating to Dubai, Switzerland, Singapore and Monaco.  The Office for Budget Responsibility’s own central scenario projects 10,800 non-domiciled resident departures a year.  A country raising the largest tax burden in sixty years, while simultaneously watching its wealthiest residents and its growth forecasts both shrink in the same eighteen-month window, is not managing a temporary rough patch.  It is managing a structural erosion of both its fiscal base and the tax base meant to service the debt these rules were built to control.

The economy had responded positively at that point.  Eighteen months of data have not vindicated that early optimism.  Growth has been downgraded twice.  Debt is rising, not falling.  Tax rises have compounded on households already absorbing the largest threshold freeze in six decades.  The interest rate on the debt now runs ahead of the growth meant to outpace it.  Reeves inherited a genuinely difficult fiscal position from her predecessors.  Eighteen months into her own framework, the numbers she set out to control are moving in the direction she promised to reverse.

---

Terence Nunis | Executive Chairman, Equinox Zenith | Author, The 1% Playbook: The Billionaire Cheat Code

Quora Answer: Can You Explain the Difference between a Strategic Withdrawal & a Strategic Retreat?

The following is my answer to a Quora question: “Can you explain the difference between a strategic withdrawal and a strategic retreat in military tactics?”

A strategic withdrawal is a planned, controlled movement of forces away from the enemy, aimed at repositioning to a stronger position or avoiding unnecessary casualties.  It demands discipline to prevent an orderly movement from collapsing into a rout, using delaying actions, ambushes, and traps to slow the enemy’s advance.

A strategic retreat is the broader category.  It can be planned or forced by circumstance, often covering greater distances, aimed at regrouping, rearming, or escaping encirclement.  The goal is preserving fighting capability for a future engagement, not merely surviving the present one.

What the Masters of War Said About This

Sun Tzu wrote in The Art of War, “He will win who knows when to fight and when not to fight.”  This single line contains the entire logic separating a withdrawal from a rout.  He also warned, “If ignorant both of your enemy and of yourself, you are certain to be in peril in every battle,” a statement of why controlled withdrawal requires accurate self-assessment, not panic dressed up as strategy.

Generalmajor Carl Philipp Gottfried von Clausewitz, the Prussian general and military theorist, addressed this in Vom Kriege: “The defensive form of war is in itself stronger than the offensive.”  He argued a well-conducted retreat could strengthen a force relative to its pursuer, drawing the attacker further from his own supply lines while the defender fell back toward his own reinforcements.  Clausewitz insisted a retreat retained its military value only for as long as the retreating force stayed organised enough to turn and fight on ground of its own choosing.

Quintus Fabius Maximus against Hannibal, 217 BC

Hannibal Barca crossed the Alps in 218 BC and destroyed successive Roman armies at the Trebia and Lake Trasimene, the latter killing an estimated 15,000 Roman soldiers.  Rome, shaken to its core, appointed Quintus Fabius Maximus dictator with special powers.  Aware his own forces could not win a pitched battle, Fabius refused to offer one.  He shadowed Hannibal’s army, staying close enough to harass supply lines and stragglers while avoiding direct confrontation, adopting a scorched-earth policy to starve the invader of local resources.  Roman critics mocked him as a coward.  His strategy nonetheless preserved Rome’s army long enough for Publius Cornelius Scipio Africanus to eventually win the decisive Battle of Zama in 202 BC.  Later writers praised him as the man “who by delaying restored the state to us,” a judgement history has never overturned.

The March of the Ten Thousand, 401 BC

Approximately 10,000 Greek mercenaries, led by Xenophon, had to withdraw after backing the losing side in the Achaemenid Civil War.  Cyrus the Younger had hired them to overthrow his brother, King Artaxerxes II of Persia.  Despite early battlefield success, the mercenaries found themselves stranded deep in hostile territory once Cyrus was killed.  Xenophon organised an epic fighting march home, defeating multiple pursuing armies along the way, a textbook strategic withdrawal maintained under near-constant threat across hundreds of miles.

Field Marshal Mikhail Illarionovich Kutuzov’s Retreat and the Burning of Moscow, 1812

Napoleon’s Grande Armée began the invasion of Russia with over 500,000 soldiers.  General Mikhail Andreas Barclay de Tolly, commanding the Russian 1^st Army, opened the campaign with a scorched-earth withdrawal, deeply unpopular with a Russian public demanding a decisive battle.  Field Marshal Mikhail Illarionovich Kutuzov, who replaced him, understood Barclay’s retreat had been correct, yet political pressure forced one major engagement at Borodino on 7^th September 1812, a brutal, inconclusive battle with enormous casualties on both sides.  Kutuzov then abandoned Moscow entirely rather than defend it, and the city was set ablaze, most likely by Russian governor Fyodor Rostopchin’s own order, denying Napoleon the winter quarters and supplies he needed.  Napoleon occupied an empty, burning capital, a hollow victory that trapped him rather than concluded his campaign.  Kutuzov rejected every peace overture that followed, continuing to harass the Grande Armée with Cossack raids as it eventually began its own catastrophic retreat.

Napoleon Bonaparte’s Retreat from Moscow, 1812

Napoleon’s retreat began on 19^th October 1812, aiming originally to force Russian compliance with the Continental System.  Over 400,000 men were lost to starvation, cold, and relentless Russian harassment, the consequence of Kutuzov’s earlier discipline meeting Napoleon’s own collapsing one.  This catastrophe shattered the aura of invincibility the Grande Armée had built over a decade, and set the conditions for Napoleon’s eventual defeat at Waterloo.  Clausewitz’s warning proved true: a withdrawal that loses organisation stops being strategy and becomes attrition against your own side.

The Long March, 1934 to 1935

Chiang Kai Shek’s Nationalist forces launched a fifth encirclement campaign against the Chinese Communist Party's base in Jiangxi, advised by German officer Johannes Friedrich Leopold von Seeckt to avoid costly frontal assaults in favour of gradual, fortified encirclement.  Roughly 86,000 Communist troops broke out on 16^th October 1934, beginning a retreat that would cover between 6,000 and 10,000 kilometres over 368 days.  The Red Army lost over 40,000 soldiers in the Battle of Xiang River alone, and by mid-December numbered barely 30,000.  At the Zunyi Conference in January 1935, this crisis elevated Mao Ze Dong to effective command, and his adoption of unpredictable, mobile routes helped the remaining force evade destruction.  Only around 8,000 of the original marchers reached Shaanxi in October 1935.  The Long March preserved the Chinese Communist Party as a fighting force and established Mao’s undisputed leadership, directly enabling the eventual founding of the People’s Republic of China in 1949.

The Allied Evacuation of Dunkirk, 1940

Over 338,000 British and French troops were evacuated between 26^th May and 4^th June 1940, codenamed Operation Dynamo, evacuating Allied forces trapped by German advances during the Battle of France.  Three of four Royal Indian Army Service Corps contingents, part of the roughly 1,700-strong Force K6, were evacuated successfully alongside British and French troops.  One RIASC contingent was captured, its soldiers becoming prisoners of war, including Jemadar Jehan Dad, who later escaped German captivity disguised as a French colonial soldier.  France’s colonial troops fared worse.  German forces separated Senegalese and other West and Central African soldiers from their white officers and massacred them in several documented locations across eastern France during the same campaign, one of the war’s least acknowledged atrocities.  Operation Dynamo remained a morale boost for the Allies despite the loss of equipment.  Equipment can be replaced.  The loss of the British Expeditionary Force’s experienced troops would have been catastrophic for the war effort that followed.

Operation Ke, the Japanese Evacuation of Guadalcanal, 1943

After months of grinding defeat on Guadalcanal, the Imperial Japanese Navy conducted a covert night-time evacuation of its surviving garrison in early February 1943, using fast destroyer runs under cover of darkness while deliberately misleading American commanders into expecting a fresh offensive rather than a withdrawal.  11,000 Japanese troops were extracted successfully, a tactical achievement in a campaign Japan had otherwise lost decisively.  Historians have since ranked Operation Ke alongside Dunkirk as one of the war’s more accomplished evacuations, proof that even the losing side of a broader campaign can execute a single withdrawal with real discipline.

The UN Retreat from Chosin Reservoir, 1950

Around 30,000 UN troops, under Major General Oliver Prince Smith, were encircled by roughly 120,000 Chinese troops in late November 1950.  Smith is credited with the line, “Retreat?  Hell, we’re not retreating; we’re just attacking in another direction.”  His forces broke out and completed a 78-mile fighting withdrawal through Hell Fire Valley and Funchilin Pass, in temperatures reaching 34 degrees below zero, to the port of Hungnam.  This was not a rout.  It was 30,000 men fighting their way out under command, inflicting 60,000 casualties on the pursuing Chinese force while suffering around 17,000 themselves.  This is the controlled, still-lethal withdrawal Clausewitz described as retaining military value.

Every example here, spanning classical Rome, ancient Greece, Napoleonic Russia, civil war China, and two different theatres of the Second World War, confirms the same principle Sun Tzu, Clausewitz, and history itself keep repeating.  A withdrawal or retreat is not a defeat.  It becomes one only when discipline collapses, as it did catastrophically for Napoleon.  Fabius, Xenophon, Kutuzov, Mao, the planners of Dynamo and Operation Ke, and the commanders at Chosin all proved the opposite: an army that retreats in good order lives to fight, and often to win, another day.

---

Terence Nunis | Executive Chairman, Equinox Zenith | Author, The 1% Playbook: The Billionaire Cheat Code