To demonstrate the massive commercial opportunities in this space, the video maps out five distinct case studies alongside the staggering global market size of the problem they solve:

• Financial Inclusion (M-KOPA): A market of 400 million people across sub-Saharan Africa who are currently “credit invisibles” unable to access formal credit.

• Conflict & Evidentiary Architecture (Hala Systems): A market of 2.1 billion people living in contexts of high and extreme fragility who need verifiable truth protocols.

• Agricultural Finance (One Acre Fund): A market of 475 million smallholder farms worldwide that require bundled financing and delivery.

• Community Health (Living Goods): A structural delivery gap resulting in 4.9 million preventable child deaths annually.

• Social Coordination (Tostan): A market of 957 million people across the fastest-growing demographic regions in West and East Africa requiring collective social infrastructure.

By applying the framework’s calculations to these vast market scales, leaders can see that solving these massive structural gaps using soft power is no longer just a charitable mission, but one of the most significant commercial opportunities of the decade. It allows leaders to see the market in the mission.

Read the report here: https://softpowerindex.lovable.app/reports

The growth engines of the last thirty years – from cheap capital to ever‑expanding trade and grant‑funded development – are either contracting, rationalising, or becoming more competitive. At the same time, the economy is being pulled toward three forces: AI, energy transition, and regionalisation, all of which run straight through hard problems. Taken together, they point to an architectural shift in how the world works. Hard problems – those that cross borders, are not owned by any one institution, and affect millions or billions of people – are becoming the most important market positions of our time.

The problems are not new. What has changed is the economics of solving them and the international environment they sit inside. And what is emerging is how leaders look at the world and decide where to spend the next decade of their attention.

Across health, financial inclusion, land rights, education, and climate infrastructure, three things are now true: outcomes can be measured with far greater precision; people and places can be reached at a fraction of previous costs; and capital can pay for verified results, not just the intent to deliver them. Taken together, that turns hard problems from obligations into structural vacancies in the global market.

Subscribe now

A Darker Macro Picture

At the same time, the wider macro picture has darkened. The World Bank’s June 2026 Global Economic Prospects presents a balance sheet of a lost decade: global growth slowing to 2.5%, the weakest pace outside outright recession in almost 20 years, and nearly half of developing economies failing since 2019 to narrow the income gap with richer countries.

By the end of 2026, one quarter of developing economies, one third of low‑income economies, and half of fragile and conflict‑affected states will be poorer than they were before COVID‑19. Private investment growth in emerging and developing economies has more than halved relative to the 2010s.

The Emerging Forces

And yet, inside that bleakness, three strong forces are gathering: AI, clean energy, and a surge in regional trade. Each is, at its core, a response to a hard problem.

AI is shorthand for a productivity crisis that standard tools have failed to fix. Even if it “under‑delivers” against the current hype, the Bank’s own estimates suggest that broad adoption would still lift global growth in the 2030s above the average of the 2000s, making it the most prosperous decade since the 1970s. Energy transition is no longer framed solely as climate policy; clean energy now accounts for two‑thirds of all global energy investment, reaching a record US$2.2 trillion in 2025, driven as much by energy security and resilience as by emissions targets. Regional trade agreements have proliferated from over 300 in 2020 to nearly 400 today, now governing around 60% of global trade and binding developing economies into denser regional architectures with clearer rules on investment, standards, and services.

If we zoom out, what we see is: growth is increasingly constraint‑driven. The constraints are hard problems.

Hard Problems: Where Potential, Profit, and Purpose Meet

Hard problems have always existed. What has changed is the economics of solving them. The collapse of the old development architecture in 2025 - USAID dismantled, UK aid redirected, bilateral flows into Africa falling by around 40% in 2026-27 - did not make the problems go away. It left a structural vacancy. The institutional model that assumed sovereign states would sustain grant‑based aid at scale, multilaterals would coordinate delivery, and funding continuity was the baseline condition for design simply no longer fits inside the remaining fiscal space.

The numbers are stark. Global military spending reached US$2.7 trillion in 2024 - ten consecutive years of growth, 2.5% of world GDP, and roughly thirteen times total global development aid. Net official development assistance fell 23% in real terms in 2025, the largest annual drop on record, and is expected to decline further, stripping away one of the last buffers supporting basic services in many low‑income countries. More than half of the world’s low‑income countries are in or at high risk of debt distress. The SDG financing gap has widened to between US$4 and 6.4 trillion per year.

Meanwhile, demand for essential services has not moved. Six hundred million children are in school and not learning. 1.1 billion people lack formal land rights. Five billion cannot access basic legal help. In 2024, 4.9 million children died before their fifth birthday, most from preventable causes that existing interventions could have avoided. At current trajectories, around 27.3 million children will die before age five between 2025 and 2030. These are tragedies. And, they are structured markets where the cost of the status quo already exceeds the cost of proven solutions.

That crossing point - the moment when solving becomes cheaper than leaving a problem unsolved - is what I call the Crossover Point. It is a calculation, not a metaphor.

In homelessness in the UK, Nicholas Pleace’s work for Crisis established that one person sleeping rough costs the public purse around GB£20,128 per year across emergency services, healthcare, temporary accommodation, and legal costs, while a successful early intervention to prevent that homelessness costs around GB£1,426 - a 14‑to‑1 ratio. The crossover has already happened. The question is whether our models, and our capital, behave as if it has.

In child survival, Living Goods’ digitally enabled community health model (the DESC model) delivers a 27-28% reduction in child mortality at US$3.09 per person per year - cheaper per outcome than what governments were already paying. That turns under‑five mortality from an unquantified tragedy into a contractable unit of survival a minister can defend to parliament.

In financial inclusion, M‑KOPA built pay‑as‑you‑go credit rails for a segment traditional finance had written off as “too risky to serve”: 400 million people across sub‑Saharan Africa without access to formal credit. By using mobile money and real‑time behavioural data, they made the loan itself the mechanism for generating credit history rather than the reward for having it, extending over US$2 billion in asset finance to seven million customers and turning “credit invisibles” into a bankable consumer segment.

Doing Good? Or Good Business for the Greater Good.

What these organisations have in common is that they hold structurally defensible positions inside hard problems where the crossover has already occurred, the chokepoints have shifted, and outcomes are contractable. They’re not driven to “do good”.

Underneath, four structural shifts are doing the heavy lifting.

First, the global aid architecture has collapsed, yet emand remains. The old model has exited the field, leaving in‑market stakeholders to look for new partners, new models, and new capital that does not carry the conditionality or timelines of the previous system.

Second, the cost of delivery has collapsed. The cost of frontier language‑model inference, the cognitive layer for tutoring and legal documentation, fell by about 98% between 2023 and early 2026, dropping from roughly US$60 to around US$0.75 per million tokens. National‑scale digital identity registration, once US$20 or more per person, now runs at under US$1 per person. Satellite imagery suitable for monitoring and evaluation has become cheap enough to replace large parts of field‑verification budgets. Deworming a child still costs around US$0.50 and yields an estimated US$169 in lifetime economic returns per dollar invested. Each of these cost curves changes which models are commercially viable.

Third, the chokepoints have dissolved. Accreditation, centralised data gatekeeping, correspondent banking, and institutional chains of trust were features of the old architecture, not of the problems themselves. DESC community health workers match or exceed clinic outcomes without being professionally registered. Legal empowerment organisations provide land tenure and contract enforcement through paralegal networks that never touch a formal court. Conflict‑zone evidence can be secured via cryptographically verified provenance, rather than institutional custody.

Fourth, outcomes have become contractable. We can put a price on a single well‑being‑adjusted life year (a WELLBY) at GB£15,900, endorsed by HM Treasury and used in social value calculations. We can price one child reading fluently at US$12 per verified outcome unit. High‑integrity carbon credits now require verified land tenure and attract investment‑grade capital. Remittance flows to Africa exceeded US$100 billion in 2024, surpassing aid; if routed through outcome contracts at current unit costs, they could finance hundreds of millions of verified health or land outcomes.

All of that runs over digital rails - UPI in India handling 16.6 billion transactions in February 2026 (US$274 billion in value), Pix in Brazil reaching 170 million users and processing more daily transactions than Visa and Mastercard combined, with the UPI architecture now licensed to Peru, Namibia, and Trinidad and Tobago. Proven models can now travel without headquarters.

So where does this leave each of us?

The World Bank warns that, barring a miracle, the 2020s will be a lost decade for dozens of developing economies, with per‑capita income in EMDEs excluding China and India not regaining its pre‑pandemic relative position versus advanced economies until after 2028. It also notes that between 2019 and 2025, the number of people facing severe food insecurity in EMDEs (excluding China) rose by around 220 million, with a further 35-70 million at risk of being added by 2028.

At the same time, it points to AI, clean energy, and regional trade as the forces powerful enough to make the 2030s a golden era for job creation and growth if we are deliberate now. Those forces do not sit on top of hard problems; they run through them.

Essentially, we get to decide our focus. And the action of our next decade.

Share

Deciding Your Next Decade

I am interested in this as a decision lens for leaders who already have the resources and ambition to build something that shifts a hard problem in a measurable way:

If you are a CEO, your exposure to hard problems is already on your balance sheet - in supply chains that depend on undocumented land, workforces without basic health coverage, or markets where financial exclusion caps your customer ceiling.

If you are an investor, you are watching models that look “too impact‑shaped” for traditional mandates but whose unit economics are moving faster than your portfolio assumptions.

If you are a founder or a former development leader, you may be circling a problem you care about, funding pieces of it, joining boards, speaking on panels - and still not have a structure that tells you whether there is a real market position inside it.

Hard problems, solved at global scale, with verified outcomes and a cost base that drops every year – that is not a gap in the market. That is the market. It is how we find the market inside the mission of doing work that is necessary, needed, and, for a number of leaders, the next focus of their career.

The lights are going out. And there’s glimmers of light shining though. Against this backdrop, the question I am sitting with, and that I invite you to sit with alongside me, is:

For the problem you care about most, has the crossover already happened - and if it has, are you prepared to treat that problem as your primary market position in the decade ahead?

Leave a comment

Further reading

• World Bank - Global Economic Prospects, June 2026
  The baseline for the “lost decade” argument, with data on slowing growth, rising debt, and collapsing private investment in emerging and developing economies.

  https://www.worldbank.org/en/publication/global-economic-prospects

• Soft Power Brief Q1 2026 – Hard Problems Become Market Positions*
  A deeper dive into the Crossover Point, structural vacancies, and the three‑calculation framework for treating hard problems as market positions.

  https://softpowerindex.lovable.app/reports

• UN - Reforms to the International Financial Architecture
  Policy brief on why the current global financing system is increasingly unfit for purpose and how to close a multi‑trillion‑dollar SDG financing gap.

  https://sdgs.un.org/sites/default/files/2023-08/our-common-agenda-policy-brief-international-finance-architecture-en.pdf

The economics of hard problems captures this shift. In domain after domain, the cost of a structural solution has fallen below the rising cost of the status quo because three things changed: we can now price a single verified “unit” of change; delivery costs have collapsed via digital rails, AI, and low‑cost infrastructure; and capital can increasingly pay for results rather than inputs. That turns problems that once depended on money and mandates into investable market positions - learning poverty, land rights, child survival, and neglected tropical diseases - and leaves a very specific set of open work for this decade: building the financing, governance, government‑adoption, geographic, and political‑economy structures that let those economics actually run at scale.

The Soft Power Index calls the threshold where this becomes true the Crossover Point: the moment the unit economics of a solution beat the compounded cost of not acting. Beyond it, persisting with the old model is fiscally irrational. Organisations that have crossed share three features: a verified outcome unit, a model that holds without exceptional conditions, and assets the system cannot easily replicate – datasets, trust architectures, and operating systems that compound over time. The Pathfinder Track maps those crossing now: 27 organisations across twelve domains that pass most, but not all, of the Index’s filters.

Subscribe now

DNDi: From R&D Scarcity to Structural Independence

DNDi has built what commercial pharma will not: a full R&D pipeline for neglected tropical diseases that affect over a billion people and generate no commercial return. Its outcome unit is a new or repurposed treatment, brought through a WHO‑endorsed pipeline, for a disease that otherwise would not attract R&D.

AI‑assisted drug repurposing has slashed the time and cost of early discovery, so the marginal cost of developing NTD treatments is now converging with commercial pharma economics for patients who were previously excluded. The gap is no longer science; it is financial architecture. DNDi remains grant‑funded and exposed to shocks like the USAID cuts, which makes “as much as needed, for as long as needed” commitments impossible.

Our World in Data: Endowing the Global Evidence Commons

Our World in Data is a global evidence commons: a place where policymakers, journalists, scientists, and students in 190+ countries see the same numbers on health, climate, poverty, and more. Its outcome unit is a high‑integrity, openly accessible indicator that decision‑makers can rely on.

The cost of running that commons is tiny compared to the cost of misinformation, policy whiplash, and duplicated data infrastructure. When USAID cut funding for the Demographic and Health Surveys, OWID’s dependence on fragile inputs was exposed at the exact moment GiveWell and others leaned on its SDG Tracker for verification. The gap is not usefulness or demand; it is the absence of a dedicated endowment to secure long‑term independence.

One Acre Fund: The Cost of Donor Dependency

One Acre Fund has built the most documented operating system for smallholder finance and productivity in the world, serving millions of farmers. Its outcome unit is a farmer completing a full bundle cycle – inputs, credit, training, and market access – with documented yield and income gains in the first season.

A bundled contract that aligns inputs, credit, and training produces food at lower total cost than fragmented subsidy programmes, with yield gains of 40–50% and repayment rates above 95% across multiple countries and climate shocks. Yet after USAID’s collapse, in May 2026, One Acre Fund announced proposed redundancies for 1,752 staff – 47% of its Kenya workforce – even as it reached 5.9 million households and delivered roughly US$600m in annual impact. The economics work; the architecture does not.

SOIL: Pricing a Public Good in a Fragile State

In Haiti, SOIL has priced something the sector long treated as unpriceable: safely managed sanitation in a fragile state. Its container‑based model delivers toilets, waste collection, and composting at a verified cost per connected household, backed first by an outcomes contract with IDB Lab and then by the world’s first government results‑based contract for container‑based sanitation with Haiti’s DINEPA. The programme has survived political collapse, gang‑controlled supply routes, and the USAID shock while meeting or exceeding its outcome targets.

The economics have crossed; the gap is financial architecture and government adoption depth. A dedicated budget line in Haiti’s WASH plan and a second government commissioning the model from a local partner would convert a fragile proof into a replicable template.

Share

Building the Market in the Mission

Across the 27 Pathfinders, the pattern repeats. Solving is now cheaper than not solving; the remaining work is architecture. The structural shifts of 2026 show hard problems affecting millions or billions of people worsening, even as the economics of solving them improve. For capital allocators, builders, and strategic partners, the question is who will close those gaps. As a result, they will claim durable positions in the hardest problems and largest global market opportunities of this decade. And, in doing so, they will build the market in the mission.

Leave a comment

A Defining Moment in Time

There is a moment, in any hard problem, when the cost of solving it structurally drops below the cost of leaving it unsolved. That moment can be calculated, not guessed. The cost of the solution falls as technology deflates and verification gets cheaper. The cost of the status quo rises as the problem compounds and institutional alternatives exit. Those are the two curves: the falling cost of solving and the rising cost of leaving it unsolved. Both curves are moving. The crossing is the Crossover Point, and it arrives earlier than a static comparison would suggest, because both curves are in motion at the same time.

For most of the last two decades, carbon could not be read this way. The cost of the status quo, what the global system was actually paying to leave emissions unpriced, was a contested number. Estimates of the social cost of carbon ranged so widely, and rested on so many disputed assumptions, that no finance minister, no procurement officer, and no corporate board could put a defensible figure on what the problem was costing. That uncertainty was the real chokepoint. It kept the entire carbon debate stuck in an argument about whether the number was real, rather than what the number unlocked.

In March 2026, Marshall Burke and colleagues at Stanford resolved that argument. The technical details are dense, yet the headline is straight-forward. A peer‑reviewed paper in Nature produced the most rigorous quantification to date: a central estimate of the social cost of approximately US$1,013 per tonne of CO₂ under conservative assumptions (2% discount rate, no impacts past the year 2100), rising to US$3,448 under continued post‑2100 growth effects at the same 2% discount rate, and to US$7,822 under Ramsey discounting through to the year 2300 (Burke et al., 2026, Extended Data Table 2). That central figure is an order of magnitude higher than the bottom‑up estimates currently used by the US Federal Government, and well above the prices in any operational carbon market on earth.

And the authors are explicit that these figures are conservative. They measure GDP loss only. They do not capture biodiversity collapse, cultural displacement, loss of ancestral homelands, sea level rise, or the full weight of extreme weather events. The paper itself states that “our damage estimates will understate the total damages associated with historical and future emissions, perhaps substantially.” Speaking to reporters at publication, co-author Noah Diffenbaugh put it plainly:

“We haven’t accounted for impacts that aren’t captured in GDP, such as loss of biodiversity and cultural homeland, and our approach also underweights some sources of climate impacts such as sea level rise and some types of extreme events. This makes our estimates conservative.”

The central figure is also, in the authors’ own framing, “much larger than recent bottom-up estimates produced by the US Federal Government” - roughly five times the EPA’s most recent figure under comparable assumptions. The unpriced cost is higher. It just hasn’t been quantified yet.

The second curve now has a peer-reviewed anchor.

Subscribe now

Pricing the Status Quo

The Burke number does something specific. It is not a bill you can hand to a single client or customer. An airline cannot tell a passenger, “we are adding a charge because your flight’s emissions cost US$1,013 per tonne,” because the damages from that tonne are shared across the whole planet and over many years, not just borne by the person buying the ticket. The same is true everywhere: the real cost of carbon is distributed across populations, ecosystems, and timescales no single buyer sees on their own balance sheet.

What the number does is anchor a trajectory that was previously a directional argument. The Burke figure is an anchor price and a liability ceiling that exerts upward pressure on every adjacent cost in the carbon stack over time.

Every cost currently rising - mandatory climate disclosure compliance under regimes like Australia’s AASB S2, the EU Corporate Sustainability Reporting Directive, California’s SB 253, rising landfill levies, insurance premium differentials on climate-exposed assets, Scope 3 push-down from major corporate buyers, ESG claim legal exposure - is rising because the underlying systemic cost is an order of magnitude higher than anyone previously priced it. Burke et al. have measured the pressure that is pushing these client-level costs up.

A few concrete figures from the paper illustrate what this trajectory looks like when translated into specific actors.

One tonne of CO₂ emitted in 1990 caused approximately US$180 in global damages by 2020. That same tonne will cause an additional US$1,840 through to 2100. The historical damage was less than 10% of the total debt.

US emissions since 1990 have caused approximately US$10.2 trillion in cumulative global damages by 2020, with about 30% (US$2.97 trillion) felt inside the US itself and 14% (US$1.39 trillion) in the European Union.

Aramco’s 1988 to 2015 emissions alone represent an estimated US$3 trillion in cumulative global damages realised by 2020 - roughly eight years of company revenue - rising more than 20-fold to US$64 trillion by the end of the century if those emissions remain in the atmosphere (Burke et al., 2026).

The distributional picture is clear. Per-capita emissions data from global inequality research (Chancel et al., World Inequality Lab) places the top 0.1% of global emitters at over 290 tonnes of CO₂ per person per year; the poorest 50% at roughly 0.7 tonnes. Applying the Burke central social cost to those figures produces approximately US$294,000 in annual damages caused per person in the top 0.1%, against roughly US$710 per person in the poorest 50%. A 400-to-1 ratio, between the populations that contribute most and the populations that bear the consequences most heavily.

As mentioned above, these numbers are conservative. They are the priced GDP component only. The full cost is larger, and not yet reducible to a single figure.

The Cost of Solving is Collapsing

While the second curve was being priced, the first curve was falling. Three cost categories - monitoring, verification, and payment - historically made carbon markets expensive to build and easy to contest. All three are now approaching marginal costs that round to zero.

Monitoring is now satellite-native. Commercial archived imagery now costs in the range of US$10–20 per square kilometre, with fresh tasking typically in the range of US$20–40 per square kilometre for sub‑metre resolution (Planet, Maxar, Airbus 2025 price tiers). A decade ago, that capability sat exclusively with institutional field missions and cost orders of magnitude more.

Verification is now cognitively cheap. Frontier‑equivalent inference costs have fallen in the range of 90–95% between early 2024 and early 2026, across multiple benchmark series (Epoch AI and Artificial Analysis tracking). The cognitive layer required to process and verify disclosure data at scale now costs almost nothing at the margin. Digital identity registration has fallen from over US$20 per person to under US$1 per person at national scale (World Bank ID4D benchmarks), making it possible to attach verified emissions data to specific transactions, assets, and producers.

Payment has scaled further than most observers have registered. India’s UPI processed 22.64 billion transactions in March 2026, a record high and a 24% year-on-year increase in volume. Brazil’s Pix has onboarded 93% of the adult population and now processes more daily transactions than Visa and Mastercard combined. Kenya, Peru, Namibia, and Trinidad and Tobago are running on licensed derivatives of the same infrastructure. Outcome-linked capital can now be routed to specific populations at costs the old architecture could not match. Burke and colleagues themselves identify this as an open avenue, noting that “well-developed opportunities for transfer payments also exist outside the international system, such as bilateral, low-cost transfer payments to the mobile phones of low-income households in developing countries.”

Monitoring, verification, and payment. Three pillars. All three have crossed the threshold where cost is no longer the binding constraint.

How to Calculate a Market Position in Carbon

Put both curves in the same frame and a specific claim becomes possible for the first time.

The cost of the status quo in carbon is priced at roughly US$1,013 per tonne globally, rising, and conservative. The cost of verifying, monitoring, and contracting against a tonne has fallen to a level where the architecture to deliver a priced, verified unit – one tonne verified and reported, one asset tracked with full material provenance, one disclosure‑ready procurement decision – is commercially deliverable at a cost well below what the system is already absorbing.

Putting a number on the second curve – the cost of leaving carbon unsolved – is what makes it possible to see when solving becomes cheaper than not solving, and to treat that gap as a market opportunity. When it is cheaper to solve, you can create offers that finance ministers, procurement officers, and corporate boards will listen to – offers that can be written into real contracts.

That is a Crossover Point. The Burke paper, published in March 2026, is what makes it calculable. The framework’s own contribution, in the authors’ words, is that the approach “aligns the financial accounting framework of L&D with the existing SC‑CO₂ framework.” That alignment is what turns a contested number into something legible to courts, disclosure regimes, and insurance underwriters – and it is what Diffenbaugh himself signals when he notes that the work clarifies “the cost‑benefit analysis of potential solutions.” The number is no longer just a damages estimate. It is a basis for calculating whether a given response is worth doing. And because the Outcome Unit – one verified tonne – is the same across every sector where carbon is embedded, the addressable market sits beyond one industry. It is the full global economy, priced at the full social cost of emissions, rising on a measured trajectory.

For the first time, a market position in carbon can be calculated. The number is large, the number is global, and the number is conservative.

Calculating the Market is the First Step, Not the Solution

An important qualification: a calculable market position is not the same as a solved problem. What the calculation establishes is that it is now commercially viable for the market to solve the problem, at a cost below what the system is already paying to leave it unsolved. That is a necessary condition. It is not a sufficient one.

Carbon remains a hard problem, and it is hard for specific reasons. It crosses borders, which means no single jurisdiction can price or enforce it comprehensively. It compounds over timescales longer than political cycles and corporate strategy horizons. It distributes damage asymmetrically, hitting populations that did not cause it and have little capacity to absorb it. It requires coordination across actors who have no prior history of coordinating and limited reason to trust each other. And it sits across public and private boundaries in a way that makes both pure-market and pure-state solutions structurally incomplete.

These are coordination problems, legitimacy problems, and trust problems. They are the class of problem that mandates, money, and market failures have a long history of failing to solve.

The open question, now that the market can be calculated, is what kind of actor will occupy the position.

What Soft Power Has Already Solved

The useful signal sits in adjacent spaces where the same class of problem is hard, cross‑border, coordination heavy, and trust dependent, and has already been crossed. Three cases are relevant references, because each demonstrates a specific dimension of what carbon now requires.

Pricing the Outcome Unit: Living Goods. Child survival in sub-Saharan Africa was, for decades, the archetypal case of a hard problem that money and mandates could not fix. Billions of dollars in donor funding, state health system budgets, and multilateral programme delivery produced activity without accountability. Inputs were tracked. Outcomes were not.

Living Goods changed the calculation. A 2021 randomised controlled trial (RCT) covering 4,500 community health workers (CHW) serving 3.6 million people in Uganda established that a compensated, digitally supervised CHW architecture produced a 30% reduction in under-five mortality and a 27% reduction in infant mortality. The Kenyan government moved the model from externally funded project to county-contracted public service. The shift was not driven by mandate. It was driven by arithmetic. The market opened because the Outcome Unit was priceable and the ratio was defensible.

Carbon has just had its RCT moment. Burke et al. have priced the Outcome Unit at an integrity standard a government buyer can defend.

Verification without intermediaries: Hala Systems. For decades, accountability for wartime atrocities depended on institutional chains of custody: human observers, NGO field missions, UN verification, state-mandated tribunals. Each required physical access, institutional backing, and months of verification before evidence could be submitted. The chokepoint was physical and institutional.

Hala Systems dissolved that chokepoint by replacing the chain of custody with a chain of provenance secured by code. Cryptographic verification removed the need for the observer. The question shifted from “who collected this evidence and can we trust them?” to “has this data been altered since it was recorded?” A court can answer the second question without an institutional intermediary. The world’s first ICC Article 15 dossier featuring cryptographically secured evidence was submitted through this architecture. The evidentiary standard is now mathematical rather than reputational.

Carbon faces the same verification problem at enormous scale. Every tonne of embodied carbon, every verified avoided emission, every traded credit requires a chain of provenance that currently depends on institutional trust, and currently collapses under sustained scrutiny. The architecture that proved itself in conflict zones, where every data point is actively contested, is precisely the architecture that carbon verification now requires.

Coordination against asymmetric harm: Tostan. For decades, international attempts to address female genital cutting and child marriage in West and East Africa relied on legal bans, state enforcement, and awareness campaigns. Each required either state capacity to enforce compliance or individual conviction to change behaviour. Neither addressed the actual problem: a family that abandoned a harmful norm alone faced prohibitive social and economic consequences that law and information could not protect against.

Tostan demonstrated that the problem was solvable, but not by hard power. It was a coordination problem. When a critical mass of community members could publicly commit together, the social equilibrium shifted and the new norm became the lower-cost choice. Over 10,000 communities across eight countries have now publicly declared abandonment. A UNICEF evaluation confirmed the change held 8 to 10 years after the intervention ended, with no further programming. The Government of Senegal integrated the model into national strategy. Ten UN agencies adopted it as a reference framework.

Carbon redistribution is the same class of problem. The payment rails now exist. UPI, Pix, and their licensed derivatives can route low-cost transfers to mobile phones of low-income households. The measurement science now exists. What is missing is the coordination layer that routes priced outcome payments from the populations causing damage to the populations bearing it. That coordination layer cannot be built by the institutions that failed to deliver climate finance under the old architecture, because they still operate under the constraints that made them fail.

It has to be built by actors who can hold credibility across jurisdictions, operate without institutional sponsorship, and verify outcomes in a way that capital can contract against. That is precisely the class of actor that has already solved the adjacent problems.

What This Means for Capital Allocation

The carbon space has just completed a transition that took child survival in Africa two decades, conflict evidence a decade, and norm change in West Africa twenty years. The Outcome Unit is priceable. The Crossover Point is documented. The payment infrastructure is operational. The verification architecture has already been battle-tested in a harder environment.

The implication for capital is more pointed than the general argument suggests. The Burke number is not a price target. The investable thesis is not “buy credits until they converge on US$1,013.” The investable thesis is that the gap between market price and true damage will be closed by litigation, disclosure, and insurance repricing, not by voluntary market convergence. That changes where capital should sit.

Even on conservative assumptions, the order of magnitude is clear. For example: human activities released roughly 53 gigatonnes of CO₂-equivalent in 2024 (all greenhouse gases on a CO₂e basis, per the UNEP Emissions Gap Report). If just 5% of that annual flow ends up under contracts, disclosures, or insurance models that reference Burke-level pricing in some form, that is a notional US$2.7 trillion of damage per year at US$1,013 per tonne. A verification-native platform that intermediates even 0.5% of that value as fees would be looking at an annual revenue line on the order of US$13 billion. The exact numbers will move; the order of magnitude will not. These are order‑of‑magnitude illustrations, not forecasts or investment advice.

That, in turn, clarifies where in the stack capital should sit. Not in the commodity layer: offset portfolios and credit generation assets in thin premium markets. Yes in the intelligence layer: MRV infrastructure, provenance systems, disclosure software, and coordination rails - the architecture that makes the gap enforceable.

And the timing signal is not the carbon price curve. It is the disclosure calendar. AASB S2 Group 1 reports are being filed now; Group 2 reporting begins 1 July 2026. California SB 253’s first Scope 1 and 2 reports are due 10 August 2026. Those venues - courts, disclosure footnotes, and insurance underwriting models - are where the Burke number becomes a financial reality. The investment horizon that matters is 2026 to 2028, not 2030 and beyond.

Share

Beyond Carbon

The calculation that opened the carbon position is not specific to carbon. It is the same calculation that opens any hard problem: define the Outcome Unit, price the cost of the status quo, measure the cost of structural solving, and identify the point where one drops below the other.

Carbon is the largest case currently calculable. It is not the only one. The Soft Power Index has mapped twelve hard problem fields where the Crossover Point has been reached or is approaching - among them antimicrobial resistance, where one institutional payment chokepoint suppressed a 28:1 return for thirty years until the NHS subscription model changed the logic in 2024; the learning crisis, where verified literacy outcomes can now be priced at US$12 per child; and access to justice, where 5 billion people are excluded from basic legal services and digital paralegal networks are starting to close the gap.

Each is calculable. Each has an Outcome Unit, a Crossover Point, and a market position waiting to be held.

Who Moves First

In carbon, cost of the hard problem can be calculated. The second curve has a peer‑reviewed anchor. The market position is defensible. The first step – calculating that the market exists – is done.

If you are a corporate, investor, builder, or operator, the move is the same: compete beyond the commodity layer of offsets and credits. Build, buy, or back the infrastructure that lets governments and companies prove, in contracts and audits, that they are structurally reducing their future carbon liability - the MRV, provenance, disclosure, and coordination systems that make Burke‑level pricing usable. That is where growing demand for solutions that reduce disclosure, legal, and insurance costs will meet a constrained supply of credible options.

What happens next depends on who moves while the position is still open. Hard problems are where the next generation of defensible market positions will be built. That is where leaders seeking positive‑sum outcomes will focus. The work is to find the market in the mission.

Leave a comment

Further Reading and Sources:

• Burke, M., Zahid, M., Diffenbaugh, N. S. & Hsiang, S. “Quantifying climate loss and damage consistent with a social cost of carbon.” Nature 651, 959–966 (2026). DOI: 10.1038/s41586-026-10272-6. SC-CO₂ scenarios and ranges drawn from Extended Data Table 2. Direct quotations from the paper text and from Extended Data discussion.

• Diffenbaugh quotations from press materials accompanying publication, 25 March 2026 (syndicated via Phys.org and Stanford research communications).

• Chancel, L. et al. World Inequality Report (World Inequality Lab), per-capita emissions data for top 0.1% and bottom 50%.

• Living Goods second randomised controlled trial (Uganda, published 2021): 4,500 CHWs, 3.6M people, 30% under-five mortality reduction, 27% infant mortality reduction.

• Hala Systems: world’s first ICC Article 15 war crime dossier featuring cryptographically secured evidence.

• Tostan: UNICEF long-term evaluation; Government of Senegal national integration; ten UN agencies adopted as reference framework.

• Technology and payments data: NPCI (UPI March 2026 volumes: 22.64 billion transactions); Banco Central do Brasil (Pix adoption); commercial satellite imagery pricing (Planet, Maxar, Airbus archives); LLM inference pricing benchmarks (2024–2026); World Bank ID4D digital identity benchmarks.

• UNEP Emissions Gap Report 2024 for 53 GtCO₂e annual emissions baseline.

• Krantz, S. Soft Power Brief Q1 2026: Hard Problems Become Market Positions. Full profiles of five organisations across financial inclusion, conflict verification, smallholder agriculture, community health, and collective norm change; scoring methodology; the three calculations framework. softpowerindex.lovable.app.

  ---

Share

Capital Without Architecture

Around five trillion US dollars in remittances have moved over the last decade. Most of this money went directly into households rather than into formal aid systems. Today, tens of thousands of former international development practitioners sit outside formal institutions with no structure to use what they know. The money to work on hard problems exists. The architecture to connect that capital to solutions does not yet exist at scale.

The assets closest to hard problems sit inside this gap. Diaspora capital already flows into households, communities, and small enterprises at a scale that often exceeds official development assistance. Operational knowledge has been built over careers in some of the most difficult environments on earth. Community delivery infrastructure already reaches the last mile at unit costs formal institutions have struggled to match. Cultural credibility and trusted relationships were built over decades and do not transfer easily to new entrants.

The old development architecture bypassed most of these assets. Centralised, grant‑based models favoured capital that could be tracked through institutional ledgers and reported against log‑frames. Assets that were relational, informal, or embedded in communities did not fit that reporting structure, so they sat outside it. When key parts of that architecture were dismantled, the assets remained in place, largely unused by formal systems and available for new operating models. A study of these soft power assets is what underpins the Soft Power Index, which tracks organisations already building on them.

Subscribe now

The Common Land

In 18th‑century England, control of common land changed hands. Communities had used it informally for grazing, farming, and basic sustenance. The land was productive; the inputs were there. What changed was the architecture: legal instruments, property rights, and capital structures that organised scattered common plots into enclosed, privately held productive units.

The people who understood this logic early built the agricultural and industrial markets that defined the next century. Large estate owners and early industrialists consolidated land, increased yields, and created the production base that industrial cities and global markets later built on. Those who continued treating common land as informal and unstructured lost access to it altogether. The model was zero‑sum: value was created and captured by the few, with control shifting away from the communities that had used the land.

There is a parallel to the current moment, although the logic can now be applied differently. Diaspora capital, institutional knowledge, and community delivery infrastructure are the common land of the hard problem space today. They are productive, present, and already used informally. The missing piece is the architecture that can turn them into clear outcome units and contractable positions. The first movers who build that architecture will hold market positions that compound; the question is who those positions are built with and for.

M‑KOPA shows how this can be done with a positive‑sum orientation. Hundreds of millions of people across sub‑Saharan Africa had no formal credit history but did have real repayment capacity and purchasing power. That latent asset sat outside formal finance. M‑KOPA built the architecture – PAYGo rails, behavioural data, and mobile money – that turned this into a contractable credit position. Millions of customers and billions in extended credit later, the position compounds with every repayment, and the underlying communities gain access to assets they could not previously finance. It is positive‑sum because the development gains come from millions of people entering and using the formal financial system.

The pattern holds across health, education, climate, and other hard problem spaces: find the latent assets, build the architecture, hold the market position – but in this case, value is created, captured, and distributed in ways that support positive‑sum growth.

What Latent Means

Latent assets are capital, capability, and trust that already exist close to hard problems but have not been structured for deployment at scale. They are under‑used, labelled as informal, or bypassed by institutional models that needed centralised control to function.

Six categories define the latent asset landscape in 2026:

1. Diaspora remittances and investment capital. Africa’s annual remittances are now around US$100 billion, exceeding official development assistance in many receiving countries. The Commonwealth diaspora carries an estimated annual investment potential of US$73.2 billion. These flows move on family and community ties and do not depend on donor budget cycles.

2. Institutional knowledge networks. When USAID ceased operations in July 2025, the systems and budgets changed. The knowledge did not. People who built community health systems, land formalisation programmes, agricultural finance models, and legal empowerment networks still hold presence, credibility, and a field‑level understanding of what works.

3. Community delivery infrastructure. Community health workers organised under the DESC model (Digitally‑enabled, Equipped, Supervised, and Compensated), along with paralegal networks, farmer cooperatives, and community facilitators, already deliver services at low unit cost and at scale. The infrastructure exists. The missing element is an outcome contract that makes it financially sustainable.

4. Digital payment rails. India’s UPI processed 16.6 billion transactions in February 2026, around US$274 billion in value. Brazil’s Pix now processes more daily transactions than credit and debit cards combined in Brazil. These rails make it possible to route diaspora capital through outcome contracts, turning a household transfer into a verifiable development investment.

5. Trust and cultural legitimacy. Trust is built slowly, lost quickly, and does not move easily between actors. A later entrant cannot buy or compress the time needed to build it. It is the factor that keeps a position in place when other conditions change.

6. Blended finance instruments. As of April 2025, 259 impact bonds were active across 40 countries, including 19 development impact bonds in low‑ and middle‑income countries. The contracting infrastructure to pay for verified outcomes already exists.

From Latent to Structured

In practice, structuring access to these assets means three things: defining a clear outcome unit that can be priced, choosing the local operators and rails that already reach households, users, or beneficiaries, and building contracts that align returns to capital with better outcomes for communities, while ensuring users are not left worse off. The details differ by sector, but the sequence does not. The organisations in the Soft Power Index are currently working this way, turning latent assets into positions that hold in health, education, climate, and finance.

Latent, and Leveraged

Taken together, these categories show that the main inputs for solving hard problems are already in the system. Capital is moving, often at household level. Talent and operational knowledge sit with people who have worked inside the problem for years. Community infrastructure and trusted relationships reach the last mile. What is missing is the architecture that can organise these latent assets into clear outcome units, contract against them, and run them at scale. The next market positions in hard problems will be built by leaders who treat these assets as their starting point, not as features of an external system.

Leave a comment

Sources and References

• Migration Policy Institute, “Global Remittances Guide,” July 2024.

• UN Office of the Special Adviser on Africa, “The Transformative Power of Digital Remittances in Africa,” August 2024.

• World Bank Data, “Personal remittances, received (current US$),” including regional data for Sub‑Saharan Africa, accessed 2024–2025.

• CitiNewsroom / MyJoyOnline, “Remittances received in Ghana last year was $4.7bn – Taptap Send Director,” February 2023; GNBCC / MyJoyOnline coverage of Ghana remittances in 2023; and World Bank remittance data for Ghana.

• Commonwealth Secretariat, “African diaspora investment spotlighted at Commonwealth roundtable,” May 2026.

• IMF, Finance & Development, “Remittances: Funds for the Folks Back Home,” and related “Back to Basics” materials on remittance use.

• Brookings Institution / Education Outcomes Fund, “Impact Bonds Snapshot – April 2025,” global database of social and development impact bonds.

• National Payments Corporation of India and associated market reporting on UPI transaction volumes and values, February 2026.

• EBANX, “Five years on, Brazil’s Pix approaches 8 billion monthly transactions,” November 2025, and related analysis of Pix user penetration.

• The Straits Times, reporting on the India–Trinidad and Tobago agreement to develop a UPI‑style payments system, October 2024.

• Public announcements and central bank communications on adoption of UPI‑like real‑time payments infrastructure in Namibia, 2024–2025.

• Living Goods and evaluation partners, randomised controlled trial evidence on community health worker impact and cost per capita, as summarised by The Life You Can Save and similar evaluators.

• BBC News, reporting on the closure of USAID and associated institutional changes, July 2025.

  ---

The Brief profiles five organisations operating at the intersection where hard problems become market positions: Hala Systems on verifiable evidence in zero-trust environments, M-KOPA on financing for the underserved, One Acre Fund on smallholder agriculture, Living Goods on community health architecture, and Tostan on community-led development.

Each profile runs the framework against the organisation’s position: the soft power assets they hold, the hard problem field they sit inside, the structural conditions that made the position available, and what it would take to hold it.

The Brief was distributed to approximately three hundred close contacts across Substack, LinkedIn, and direct email in March 2026. It is the first in a quarterly series.

Download the Q1 Soft Power Brief

Leaders who care about solving a hard problem that affects millions or billions of people often start with the same questions:

Who has the money?
Who has the mandate?
How do we get close to them?

There’s an alternative:

Begin with a calculation that tells you whether the economics support you building a commercially viable solution.

Subscribe now

Starting from money and mandates consumes significant time and reputational capital in optimising for access. Strategies end up designed around budget cycles, institutional priorities, and frameworks that were not built around the health, education, financial access, or climate outcomes that actually matter.

The result from this default starting point is familiar: meetings, strategies, and pilots, but very few positions that can hold without strong relationship management and new funding. The cost is years of leadership attention diluted across work that was never structurally set up to endure.

This is grounded in a study of organisations across the Soft Power Index. The consistent finding: solutions built on credibility, verified outcomes, and architectures that stand on their own economics outperform those built primarily around access to funding and institutional mandates.

The alternative is to begin with two numbers in the same sentence: the cost of delivering one unit of your outcome, and the cost the system already pays for one unit of the status quo.

In field after field, that comparison shows that solving is now cheaper than leaving the problem unsolved. When that crossover has already happened, you are no longer begging for mandates; you are looking at a structural opportunity.

When it has not, the calculation tells you to redesign or to walk away before you commit the next decade to something that cannot stand on its own economics.

Discernment is letting that calculation narrow the field. It gives you permission to say: this problem, in this configuration, is not yet where I should build.

Determination is what comes after, when the numbers say the economics hold and you decide to commit to one clear outcome on your own terms, not as a guest in someone else’s architecture. That determination forces everything else to follow: time, attention, talent, and capital align around a position built to endure

In this short video, I walk through this shift - from defaulting to money and mandates, to beginning with a calculation that tells you whether the economics support you building a commercially viable solution.

What changes for you if, before your next initiative, you commit to knowing whether the economics support you building a commercially viable solution?

Leave a comment

Share

Most hard problems are stuck not because nobody cares, but because the people closest to them are waiting for permission that will not come. The work is to recognise that the permission is no longer the constraint, and to act on what is.

Three calculations decide whether to move: whether the problem has crossed the point where solving it is structurally cheaper than the status quo, whether the chokepoints that used to stop you have dissolved, and whether the model holds without the conditions that made it possible to start.

This work results in finding - and holding - a market in the mission.

Recent articles:

Most people with the resources and the ambition to build inside one have already done the market analysis. They know the scale. They have a view on the technology. They understand, at least in outline, what a solution would require.

What they often have not yet done is map the chokepoints.

A chokepoint is the specific point in a system where progress stalls - not because the problem is unsolvable, but because an institution built the rules when the economics were different. That distinction determines whether five years of effort produces a market position or a very expensive lesson.

Structural chokepoints belong to the problem itself. Physics that will not bend for your business model. Biology that sets hard limits on dose and risk. The deep complexity of human behaviour at scale that no single product can simplify. These are the parts of the landscape you have to redesign around, not vote away.

Institutional chokepoints belong to the institutions that last shaped the field. A reimbursement model designed for a different era. A licensing framework built around a delivery system that no longer exists. A verification standard inherited from an institution that has since exited the field. These can dissolve when the institution changes, reforms, or leaves. Often, they have already dissolved. The model has yet to update because nobody ran the calculation.

This is the pattern playing out across eleven hard problem fields simultaneously. It is now economically viable to solve these problems. For years, institutions held the positions closed. As those institutions change or exit, the positions open. Capable people are still circling because they are running their assessment against the old map.

The Chokepoint Map runs on a short set of questions. They take the circling phase and give it a direction.

Subscribe now

The Pattern Across Eleven Fields

Across these hard problem fields, the same pattern appears.

The cost curves moved - in most cases, years ago. The cost of the status quo is now measurably higher than the structural solution cost across field after field. That is the Crossover Point, operating at scale.

The economics have crossed. The barrier keeping capable people out sits between that crossing and the market position - not in the problem itself. In most cases, it belongs to an institution that has already changed.

Three examples, briefly.

Energy access. Battery pack prices fell 93% since 2010. Solar PV now costs around US$0.04 per kilowatt-hour. Off-grid solar is the most cost-effective route to electricity for 41% of the people still without access by 2030. The infrastructure wall is gone. What remains is permitting friction and grid interconnection requirements. Institutional residue, held in place by rules written for a different cost structure.

Financial exclusion. India’s JAM Trinity - combining Aadhaar identity, Jan Dhan accounts, and UPI payment rails - turned 1.3 billion people from unreachable to transactable at near-zero marginal cost. The identity and access chokepoint that blocked an entire generation of financial inclusion models dissolved with it. GDP in mobile-money countries was US$600 billion higher in 2022 than it would have been without mobile money.

Legal exclusion. 1.4 billion people have unmet civil and administrative justice needs globally. Legal problems cost individuals an average of 1.7% of GDP in lost income and related burdens. Online dispute resolution systems now operate at roughly twelve times lower cost than comparable traditional court processes. The gatekeeper chokepoint is moving. The position is opening.

These are the same story in different fields. The institution held the model in place. The institution changed. The position is available to whoever calculates it correctly and builds with the right assets.

AMR: The Clearest Case

Antimicrobial resistance (AMR) is the sharpest example in the data because the chokepoint held for thirty years against an ROI that was always visible.

The economics always supported the case. A 2024 analysis by the Centre for Global Development estimated that improving innovation and universal access to high-quality AMR treatment would cost approximately US$63 billion per year and deliver a global return on investment of 28:1 by 2050. The status quo cost, hospital costs attributable to resistant infections, runs at US$693 billion annually. Without intervention, AMR costs are projected to reach US$159 billion per year by 2050 - more than double the US$66 billion it costs annually today.

The economics crossed decades ago. The solution architecture existed. The market position was real.

New antibiotic development stalled anyway.

The reason was a single institutional chokepoint: volume-based reimbursement. Antibiotics are most valuable when used sparingly, to prevent resistance. Sparse use meant low sales volumes. Low sales volumes meant developers could recoup R&D costs only under exceptional circumstances. Correct clinical behaviour. Economically unattractive commercial behaviour. Capable capital looked at the field, saw the chokepoint, and redirected elsewhere.

The barrier was a market design failure. An incentive structure built for a different era of medicine, held in place by institutional inertia, that made a 28:1 return on investment commercially irrelevant.

In August 2024, the NHS moved its antimicrobial subscription model from pilot to full operational procurement. Payments are now delinked from sales volume. Developers receive a fixed annual fee, up to £20 million per drug, regardless of how many prescriptions are written. The United States PASTEUR Act, reintroduced in 2026, proposes fixed annual payments of US$75 million to US$300 million per novel antibiotic, projecting a 125:1 social return on investment.

The chokepoint dissolved. The position opened.

Shionogi and Innoviva Specialty Therapeutics are among the organisations that moved into AMR development specifically because the reimbursement architecture changed. The capital did not change. The science did not change. The market design changed and the position became available.

Thirty years of dissolved economics. One policy change. The market position that was always there became buildable.

How to Spot a Chokepoint

Three tests identify whether you are looking at a chokepoint at all.

If this point is blocked, does the rest of the model stall?

If it moved tomorrow, would your unit economics and delivery model already work?

Does the leverage sit in one decision, in one place, by one class of actor?

Once you have confirmed the chokepoint, establish what kind it is:

• Does this barrier persist regardless of further cost reduction? Structural.

• Is there a named rule, code, or convention holding it in place? Institutional.

• Can you point to a jurisdiction or sector where it has already moved on the same facts? Institutional.

The Eight Chokepoints

The Chokepoint Map covers eight categories. Each appears in the data across the eleven fields. Each has dissolved examples. The test for every barrier is the same: does this constraint belong to the problem, or to the institution that last tried.

Incentive and market design. The commercial model makes solving the problem economically unattractive even when the ROI is clear. AMR is the anchor: volume-based reimbursement made correct clinical behaviour commercially irrational for thirty years. The NHS subscription model changed the payment logic. The field became investable on the same science and the same need.

Measurement and verification. No verifiable Outcome Unit means no procurement decision. The chokepoint is not the data - it is the measurement architecture that makes the solution contractable. Where that architecture exists, capital follows.

Identity and access. The end beneficiary cannot be reached at a cost the model can sustain. India’s JAM Trinity dissolved this at scale across 1.3 billion people. Anyone who has built a business will recognise this pattern. Getting customers verified, onboarded, and reached at a viable cost were chokepoints in their first chapter too. The same dissolution is underway across multiple hard problem markets.

Infrastructure. In several fields, the physical and digital infrastructure wall has gone. What remains is institutional residue: permitting friction, outdated grid codes, absent common protocols. These look like infrastructure chokepoints. They belong to the institution.

Financing and risk structure. The capital exists but sits in the wrong structure. PAYGo solar converted a US$200 upfront barrier into a US$10 entry point. Parametric insurance sent US$30 million to Caribbean governments within two weeks of Hurricane Beryl, triggered by wind speed data. The chokepoint was the risk structure. The risk structure was institutional.

Regulatory. Yesterday’s rulebook governs today’s economics. The test is direct: does this constraint exist because of the problem, or because of the institution that last attempted to solve it. Many regulatory barriers have yet to update to match the current cost structure. Most will.

Trust and legitimacy. The cost curves have moved but the human confidence layer has yet to follow. Trust is earned through proximity, not purchased through scale. The organisations that hold it - built over years inside the problem - hold a position that capital alone cannot replicate. These are soft power assets. They are what converts a calculated market position into a defensible one.

Workforce and gatekeepers. The previous model required accredited specialists at ratios the system could not sustain. More than 169 million Americans live in a mental health professional shortage area. The technology to offload large portions of the work now exists. The chokepoint is the professional architecture held in place by credentialing and habit - not by the requirements of the problem itself.

For someone who has navigated complex systems at scale - whether building and selling a company, allocating capital across markets, or stewarding wealth across generations - several of these will be immediately recognisable. The chokepoints have different names but the same structure. The difference in hard problem fields is institutional age: the barriers are older, more embedded, and in many cases already dissolving without anyone having updated the map.

Share

How to Turn the Diagnosis into a Decision

The Chokepoint Map runs on two questions. They are quick to ask, slower to answer well.

Is this barrier intrinsic to the problem, or does it belong to the institution that last tried?

If it belongs to the institution: has that institution already exited, reformed, or changed the conditions?

A third check refines both.

Can you point to a country, sector, or programme where this same barrier has already shifted under similar conditions?

If yes, the chokepoint is institutional. If the answer is nowhere, look harder before assuming the constraint is structural.

Most models encounter a mix. A financing structure chokepoint that has dissolved. A verification chokepoint that has yet to move. A regulatory barrier that is institutional but still active. The Chokepoint Map produces a precise picture of where the friction actually sits. That precision is the basis for deciding where to direct the next five years.

Deploying effort against a constraint that no longer exists is one of the most expensive mistakes a well-designed model makes. Assuming a constraint is structural when it belongs to the institution is the second. Both are common. Both are calculable.

Hard Problems are Market Positions

The Crossover Point tells you whether the economics support a position. The Chokepoint Map tells you whether the path is clear.

Together they answer the question that matters. Is this worth the next decade, and can I build something that holds.

The AMR story is the answer in compressed form. The ROI was always there. The chokepoint held the position closed. The institution changed. The organisations that calculated it correctly moved first.

That sequence is repeating. Energy. Finance. Health. Legal systems. Water. Education. The positions vacated by exiting institutions go to whoever arrives with the right calculation and the right assets.

Hard problems are market positions. The Chokepoint Map is how you establish whether yours is clear to build into.

Leave a comment

The Business Case Calculation

The Business Case Calculation applies the Chokepoint Map to your specific problem and position. Ninety minutes. One page. Delivered within 48 hours.

softpowerindex.lovable.app/work-together

The Q1 Soft Power Brief applies the Three Calculations to five organisations that have already reached the Crossover Point. Each is building a global market position. Available at https://softpowerindex.lovable.app/reports

Sources

• Centre for Global Development, Forecasting the Fallout from AMR, 2024
  https://www.cgdev.org/publication/forecasting-fallout-amr-economic-impacts-antimicrobial-resistance-humans

• WHO, Antimicrobial Resistance Factsheet, 2023
  https://www.who.int/news-room/fact-sheets/detail/antimicrobial-resistance

• WHO / Global AMR R&D Hub, Incentivising the Development of New Antibacterial Treatments, 2024
  https://cdn.who.int/media/docs/default-source/2021-dha-docs/g7progress_2024_hub_who.pdf

• NHS England / NICE, Antimicrobial Subscription Model, August 2024
  https://www.nice.org.uk/what-nice-does/life-sciences-how-to-get-your-product-to-market/a-new-model-for-evaluating-and-purchasing-antimicrobials

• PASTEUR Act reintroduction, US Congress, 2026
  https://amr.solutions/2026/02/12/pasteur-v2-0-reintroduced-in-the-us-congress/

• World Bank ESMAP and GOGLA, Off-Grid Solar Market Trends Report, 2024
  https://www.gogla.org

• IRENA, Renewable Power Generation Costs in 2023
  https://www.irena.org/Publications
  https://reglobal.org/renewable-power-generation-costs-in-2023-irena/

• World Justice Project, Justice Data Graphical Report II, 2023
  https://worldjusticeproject.org/our-work/research-and-data/wjp-justice-data-graphical-report-ii

• GSMA, Mobile Money’s Impact on Economic Growth in Five African Markets, 2025
  https://www.gsma.com/solutions-and-impact/connectivity-for-good/mobile-for-development/blog/mobile-moneys-impact-on-economic-growth-in-five-african-markets

• IMF Finance and Development, The India Stack is Revolutionising Access to Finance, 2021
  https://www.imf.org/external/pubs/ft/fandd/2021/07/india-stack-financial-access-and-digital-inclusion.htm

• UNEP, Adaptation Gap Report, 2023
  https://www.unep.org/resources/adaptation-gap-report-2023

Instead of just looking at the massive scale of a problem, leaders and innovators need to compare two distinct economic curves:

• The First Curve (Supply): The actual, verifiable cost of delivering a unit of change.

• The Second Curve (Demand): The compounding, systemic cost society is already paying to maintain the status quo (e.g., emergency services, lost productivity, welfare).

When you can prove that your cost of solving is lower than the cost of the status quo, funding the solution stops being a philanthropic plea and becomes a highly logical procurement decision for governments and capital allocators.

Two striking examples of this principle in action:

• Homelessness in the UK: Research showed that leaving a single person on the street costs public services £20,128 annually, whereas the early intervention required to prevent it costs just £1,426 - a massive 14-to-1 ratio.

• Community Health in Africa: An organization called Living Goods bypassed traditional, top-heavy health institutions by using a digitally-enabled, community-level health worker model. They successfully reduced under-five child mortality in Uganda by 28% for a verifiable cost of just US$3.09 per person per year.

Ultimately, to reach this Crossover Point, an organization must define a single, verifiable “Outcome Unit” with a clear cost attached (e.g., “one child covered” or “one household with formal land tenure”). Once a model proves it can achieve this outcome consistently without relying on its founders or optimal conditions- a threshold Sophie Krantz calls the “Voltage Test”- it has the potential to scale globally.

Read the original article here:

https://www.sophiekrantz.com/p/calculating-the-second-curve

Almost none have calculated the cost of solving it against what the system is already paying to leave it unsolved. Those are different exercises. The first produces a budget. The second produces a position.

That second number, what it actually takes to deliver a verified unit of change at a price the system will pay, reframes everything. It tells you whether your solution is priced for a charity or for a contract. It tells you whether the economics have crossed. And it is the number that makes your position defensible to a government buyer, a capital allocator, or a serious investor.

This is the Crossover Point: the moment at which solving structurally becomes cheaper than leaving unsolved. Two curves, moving in opposite directions. The organisations that calculate where they cross move while others are still funding the old model.

You have priced the first curve. This edition is about the second.

Subscribe now

What is the Crossover Point?

The cost of the solution, its unit economics, is a supply-side question. The Crossover Point calculates the demand-side: what the system pays when the problem stays unsolved, and when the cost of solving drops below that.

A homelessness calculation is the clearest example. Research by Nicholas Pleace for Crisis UK established that the annual cost to public spending of a single person sleeping on the streets is GB£20,128: across emergency services, healthcare, temporary accommodation, and legal processing. The cost of the early intervention that prevents it: GB£1,426. This is a 14-to-1 ratio. The economics were always there. The solution architecture was not.

That ratio is the second curve made visible. Once you know what the system is paying on one side, the cost of solving on the other becomes a procurement decision rather than a philanthropic one.

Both curves are moving. The cost of solving falls as technology deflates and the marginal cost of each additional verified outcome approaches zero. The cost of the status quo rises as the problem compounds and institutional alternatives contract.

Knowing your solution is cheaper than the alternative is the starting point. The calculation is what converts that into a number a government buyer, a capital allocator, or an investor can place next to other options on a spreadsheet. That is what makes the position defensible.

Living Goods: Calculating the second curve in community health

For decades, the institutions with the mandate to solve community health in sub-Saharan Africa had the resources, the political access, and the stated commitment. The results were not proportional. By the early 2000s, child mortality in Uganda remained among the highest in the world. Clinic infrastructure was underfunded and unevenly distributed. Volunteer community health worker programmes operated without supervision structures, without reliable supply chains, and without the data systems that would allow anyone to know what the investment was purchasing.

Health budgets funded activity. They did not buy outcomes.

The hard power approach, involving centralised government systems and large multilateral health programmes, had demonstrated its ceiling. What it could not do was reach the last mile at a cost the system could sustain. That gap was structural, and it was large.

Living Goods entered with a different set of assets: a commercially-trained operating model, community trust built through local health workers known by name in the households they served, and a technology infrastructure that made supervision and verification possible at scale. These are soft power assets. The kind of leverage institutional health systems find hard to replicate, because the trust is earned at the community level and the operating logic runs on local knowledge rather than centralised authority.

The DESC model, meaning Digitally-enabled, Equipped, Supervised, and Compensated community health workers, put those assets inside a measurable architecture. Every element was designed to be verifiable and replicable without depending on the founding team’s presence or the next grant cycle.

The size of the problem, or the TAM, was never the question. Every health ministry knew the scale of the problem; 1.8 billion people globally living in communities where primary healthcare remains inaccessible or ineffective. The second curve, what verified community health coverage would actually cost against what the system was already paying, had never been calculated.

Then Living Goods ran the numbers.

Two gold-standard randomised controlled trials (RCT) in Uganda, in 2013 and 2021, produced consistent results: a 27-28% reduction in under-five mortality at a cost of approximately US$3.09 per person per year for full community health coverage.

The second trial is worth noting separately. The 2013 RCT ran in carefully selected areas under close management. The 2021 trial ran across 13 districts, 500 villages, and over 12,500 households, a far broader and less controlled environment. The 28% mortality reduction held. The model did not depend on optimal conditions to produce its result.

At that point the comparison became unavoidable. This is what a procurement decision looks like in practice: governments placing two numbers side by side, ongoing spending with uncertain outcomes against a defined cost per verified reduction in mortality. County-level contracts in Isiolo and Kisumu in Kenya followed. The model moved from externally funded project to government-contracted public service. Living Goods no longer needed to be in the room for it to operate.

Each government contract generates the verification data that makes the next contract faster to close. The marginal cost of each additional verified outcome falls with every cycle. The architecture scales because the Outcome Unit is fixed and the cost is known.

US$3.09. The institutions with the mandate tried for decades. Living Goods calculated the second curve and built from there.

Four questions to run on your own problem

The Crossover Point exists in every field where a structural solution is cheaper than the compounding cost of the status quo. The question is whether you have calculated the second curve. To calculate this, ask:

1. What is the single verifiable unit of change your model produces?

One child covered. One household with formal land tenure. One farmer completing a full input-to-market cycle. If you cannot write it in one sentence with a cost attached, you have a mission with activities. The Outcome Unit is the bridge between ambition and a number a buyer can act on. Without it, the second curve has nothing to anchor to.

2. Has the cost of solving dropped below the cost of leaving it unsolved, and can you prove it?

What is the system currently paying: in health costs, foregone productivity, enforcement, welfare transfers, or lost tax revenue, for each unit that does not reach the desired outcome? Put both numbers in the same sentence. If the cost of the status quo is not materially higher than the cost of your solution, the model, the buyer, or the problem entry point needs reexamining. The second curve exists; the question is whether you have found the right cross-section of it.

3. Which barriers are structural, and which belonged to the institution that last tried?

The constraints that stopped the previous model are not necessarily intrinsic to the problem. Many belong to the institution that attempted it: professional accreditation requirements, centralised data infrastructure, or headquarters sign-offs. When the institution exits, those constraints exit with it. Deploying effort against a constraint that no longer exists is one of the most expensive mistakes a well-designed model makes. The Chokepoint Map, which is the subject of the next edition, is the tool for making that distinction before committing five years to the wrong problem.

4. Would this still work if you removed the founder, the grant, or the relationship that made it possible?

If the answer is no, you have a project. Living Goods passed this test twice. The first RCT ran in selected areas under close management. The second ran across 13 districts, 500 villages, and over 12,500 households. The mortality reduction held at 28%. That is the Voltage Test: does the model keep its result when it moves beyond optimal conditions? When the Kenyan government contracted the DESC model as a funded public service, it passed a third time. The incentive structure, the verification mechanism, and the repayment architecture are embedded in the design, held there by the system rather than by exceptional conditions. A model that depends on its founder to function has not yet calculated the second curve. It has calculated the founder.

Share

What the calculation reveals

If you can answer all four questions with numbers, you have the foundation of a market position. The second curve is calculable, the Crossover Point is behind you, and the work is execution.

If two or three answers are clean and one is not, you know exactly where to direct the next six months. The gap is specific, which means it is closeable.

If the cost of the status quo is lower than the cost of your solution, the economics have not crossed. A different model, a different buyer, or a different entry point may change that. The calculation tells you which. That is more useful than another year of conviction.

There is a fourth outcome the calculation produces, and it is the one most leaders at this stage have not considered. When the Crossover Point is documented and the Outcome Unit is verified, the problem stops being local. The same economics that make the model viable in one market make it contractable in twenty. The cost structure that works in Kisumu works in Kampala, Dhaka, and Nairobi for the same reason: the second curve has a trajectory that does not respect borders. Hard problems at this scale are global by definition. The Global Market Scale can be calculated.

Capital is necessary and rarely sufficient. It cannot buy the trust, the operational knowledge, or the credibility that makes a solution replicable across markets. Those assets are earned inside the problem, over time, at the community level, and they are the ones the system cannot replicate. The organisations that have moved from single-market proof to global architecture have done so by combining a verified, low-cost Outcome Unit with three assets: relationships built at the community level, operational knowledge earned inside the problem, and the credibility that comes from having stayed when institutional alternatives left. These are soft power assets. They are what converts a defensible market position into a global one.

Most leaders at this stage have the ambition to solve a hard problem, the evidence it exists, and, more often than they recognise, and the assets to build a solution at global scale. What tends to be missing is the calculation that makes all three legible to a buyer, a funder, or a partner: the second curve, priced and documented, in a form that produces a decision rather than a conversation.

Leave a comment

The Business Case Calculation

The Business Case Calculation is a 90-minute working session. The output is a single page: your Crossover Point, your Outcome Unit, and the three calculations that establish whether your model has a defensible market position. Delivered within 48 hours.

If you are working on a hard problem and the second curve is the calculation you have not yet run, this is where that work begins.

Details here: softpowerindex.lovable.app/work-together

Next edition: the Chokepoint Map; which barriers are structural, which are institutional, and how to tell the difference before committing five years to the wrong problem.

Further Reading + Sources:

• Pleace, N. / Crisis UK. At What Cost? 2015

• Living Goods randomised controlled trial data, Uganda 2013 and 2021. Child mortality figures: UNICEF, 2024.

• The full Living Goods case study, including full Crossover Point calculation is in the Soft Power Brief Q1.

Download it here.

Minouche Shafik - former Deputy MD of the IMF, former President of Columbia and the LSE - published Development Without Aid last year in Project Syndicate. Shafik argues that new conditions demand a new global development architecture. Many leaders agree. The Q1 report starts where her piece stops: with what that shift actually looks like in operating models, unit economics, and market positions.

The stress test arrived in January 2025. Tens of billions of dollars in development funding withdrew overnight. It surfaced a specific set of models that had never been dependent on that architecture in the first place. In 2026, that distinction is now a market position.

Subscribe now

One Acre Fund is a clear example.

Across East and Southern Africa, One Acre Fund has maintained loan repayment rates around 95-99%, with organisation‑wide repayment typically 97% or higher, and farmers increasing their farm income by roughly 35–40% on supported land even after all borrowing costs are repaid (Comprehensive Impact Report, 2023 Financial Performance, Our impact, Focusing Philanthropy overview).

Its seasonal loan model lets farmers repay over the harvest cycle, so repayments help finance the next season’s inputs - economically, it behaves far more like a lean financial services platform than a traditional charity (Flexible Repayment at One Acre Fund, Comprehensive Impact Report).

That is the edge the Brief is interested in: organisations that used three calculations to create structural advantages their competitors cannot copy.

They know the Crossover Point where their solution is now cheaper than the system’s cost of doing nothing - and can price against that gap.

They’ve mapped Chokepoints and routed around institutional gatekeepers, recovering distribution, data, and pricing power.

They’ve designed for the Voltage Test, so a competent team in another geography can run the model without the founder - which is what makes it financeable at scale.

This logic has always been available in theory. What has changed is that it is now executable. Solution costs have fallen off a curve, including:

• High‑resolution satellite imagery that once cost hundreds of dollars per square kilometre is now available for a few dollars per km² from commercial providers such as SkyWatch and GeOpera.

• AI‑enabled legal workflows are cutting document and review costs by well over 40–60% in real deployments, turning what used to be six‑figure processes into a fraction of the original spend (AI legal discovery cost calculator, cost–benefit analysis of AI legal document comparison).

• Digital identity verification, once priced in the tens of dollars per manual check, is converging toward sub‑dollar, mass‑scale verification as the market matures (Juniper Research on digital identity verification costs).

Simultaneously, the financing mechanisms that can contract against verified outcomes - development impact bonds and other results‑based instruments, results‑based health financing, asset‑based revolving loan structures for smallholders, and carbon markets that pay against measured emission reductions - are more mature and more accessible than at any previous moment. In most hard problem spaces, the Crossover Point has already been crossed, and each year the solutions become more commercially viable as costs fall further and outcome‑based capital scales. The window is open. It will not stay open indefinitely.

Share

The Soft Power Brief Q1 Report is asking what this logic would look like in your balance sheet - or in the decision you have been postponing because the numbers were not yet clear enough to move.

Once you can show, in numbers, that fixing the problem is cheaper than maintaining the status quo, you are no longer pitching a mission. You are presenting an investment case - to commercial capital, to development finance, to governments under pressure to do more with less. Or you are presenting it to yourself - as proof that this is where your next five years belong.

If you are a leader working on a hard problem and want the Q1 Brief, download it here.

If you want to work through the three calculations on your specific model, that conversation also starts with a reply.

Leave a comment

This is Edition 1.

Briefing at a Glance

The Problem: AI is displacing white-collar workers faster than institutions can respond. The workers most at risk are mid-career professionals - such as paralegals, accountants, financial analysts, and administrative coordinators - whose skills were valuable six months ago and are depreciating faster than any previous technology transition has produced. [0]

The Anchor: Raghuram Rajan, former IMF Chief Economist and former Governor of the Reserve Bank of India, published Are We Facing an AI Nightmare? in Project Syndicate on 10 March, 2026. His argument: there are no easy public-policy responses to large-scale but not universal technological unemployment. It’s rigorous, unsentimental, and stops exactly where the business decisions to act begin.

The Gap: Rajan does the macroeconomic work. The Soft Power Brief does the business case.

The Calculations

1. The Crossover Point: a technology-enabled transition model is already cheaper than 60 to 90 days of institutional unemployment support in most high-income displacement scenarios. [1]

2. The Chokepoint Map: the route to scale a solution does not require institutional permission. It already exists - for now.

3. The Voltage Test: most reskilling programmes are built to outlast their funding cycle, not their founding team. That is why many fail when it matters most.

The Market Opportunity: The three calculations produce a specific finding: a leader with trusted proximity to displaced workers, independence from platform employment models, and the freedom to define success on the worker's terms holds a comparative advantage no government programme and no platform can replicate. The window is 18 to 36 months.

Subscribe now

The Problem

Raghuram Rajan’s analysis is a rigorous framing of the AI displacement problem. It maps scenarios without the false comfort of positive‑sum institutional solutions and names the failure modes precisely: government response will be slow, fragmented, and calibrated to political pain thresholds rather than worker needs, while technology platforms capture most of the transition value, further entrenching winner-takes-all outcomes. The problem is structurally hard because it does not belong to any single government, sector, or jurisdiction; it cuts across borders, industries, and professions in ways no one actor is mandated – or incentivised – to solve end‑to‑end.

The workers most at risk are not low‑wage workers who can be absorbed into service‑sector employment. They are mid‑career professionals with mortgages and specialised skills: paralegals, accountants, financial analysts, administrative coordinators, and diagnostic or decision‑support roles in medicine. These roles are being replicated by AI faster than policy can respond – and faster than the workers affected can self‑direct their way out. The data is already in the ledger: middle‑management postings have fallen sharply, professional‑services openings are at decade lows, and roughly 40% of white‑collar job seekers in 2024 did not secure a single interview. These are not projections. They are current conditions. [0]

Rajan’s core argument: there are no easy public‑policy responses to large‑scale but not universal technological unemployment. His Goldilocks scenario – AI rollout not too fast, industry not too oligopolistic – is the right frame for what institutions should be hoping for. It is not instructive for a business leader or founder deciding what to build.

The gap the analysis doesn’t reach is specific: who builds the transition architecture that serves displaced workers before platforms consolidate around credential infrastructure they own – and how it is financed without waiting for government programme cycles that will arrive too late. One design constraint any serious transition architecture must account for is the risk of double displacement. A worker reskilled into a role that is itself automated within 12 to 18 months has not been transitioned; they have been cycled. Evidence from existing transition programmes suggests that models oriented toward verifiable, transferable skills rather than role‑specific retraining are more durable against this risk. This brief shows where the gap is and what the structural requirements are; the specific design belongs to the leader with trusted access to the people the problem affects.

This is therefore not a policy question but a soft power question – for leaders with trusted proximity to the people the problem affects, independent enough to name what platforms are doing to portable worker value, and structurally positioned to move without institutional permission.

Hard power attempts to solve this problem are instructive. The US Trade Adjustment Assistance programme was repeatedly defunded, restructured, and ultimately allowed to expire in 2022. The UK’s National Retraining Scheme was announced in 2018 and quietly abandoned by 2020 without meaningful scale. Australia’s JobActive programme went through multiple reinventions, each redesigned around incoming political priorities rather than persistent worker need. These programmes did not fail for lack of commitment but because of three structural features hard power cannot overcome: they were jurisdiction‑bound in a problem that crosses borders and sectors; their definition of success was owned by funders and political cycles rather than workers; and they were designed for a political moment rather than the decade of displacement that followed. When the moment shifted, the programmes shifted with it. The workers were left where they were. [2]

Rajan’s conclusion creates the starting line for this brief: governments will move too slowly and platforms will move in their own interests, which means the decisive architecture will be built in the space in between. His analysis tells us why the system will fail displaced workers. That is where we begin – by defining, in falsifiable terms, what it would mean to succeed for a single worker caught in that transition, and then running the three calculations against that definition.

The Outcome Unit

What we are calculating towards.

Before any of the three calculations can be run, one requirement comes first: a precise, binary, verifiable definition of what solving this problem actually means. Not reskilled. Not enrolled in a programme. Not placed. Those are programme‑owned labels. They cannot be independently verified or financed against, and they let funders decide what counts as success – which is exactly when a leader’s independence is most at risk.

The Outcome Unit for AI displacement, an example:

One worker with portable, verified credentials recognised across multiple employers and platforms, in employment or self‑employment generating income above their pre‑displacement baseline, confirmed by independent income data at 18 months (via open‑banking APIs, payroll integration, or tax data, depending on jurisdiction).

Every word in that definition is load‑bearing. Portable means the credential is not locked inside a single platform’s ecosystem. Verified means independently confirmed, not self‑reported. Recognised across multiple employers means the market, not just the programme, has validated it. Income above pre‑displacement baseline means genuine recovery, not just activity. Confirmed by independent income data means the outcome is falsifiable – it either happened or it didn’t, and someone other than the programme can check.

One important boundary: this Outcome Unit applies to skills‑based roles, where competency – not statutory licence – determines employability. Paralegals, financial analysts, administrative coordinators, and many diagnostic support roles fall in this category. Statutorily licensed roles – qualified lawyers, certified public accountants signing audits, licensed medical practitioners – face legal constraints that portable credentials cannot bypass. The model described here targets the larger, faster‑growing category where the gate is a hiring filter, not a legal requirement.

That definition is what the Crossover Point calculates against, what the Chokepoint Map routes toward, and what the Voltage Test asks whether the model can produce without the founder in the room. Without it, none of the three calculations can produce a number. With it, all three become precise.

Share

The Three Calculations

1. The Crossover Point

What it is: the moment a structural solution to a hard problem becomes cheaper than the system’s cost of doing nothing. It rests on Wright’s Law – costs fall by a predictable percentage with every doubling of cumulative production – which means the crossing point is calculable in advance, not guessed after the fact.

The calculation creates two strategic positions. If the crossover has already happened, the question is not whether to act but why the institution has not adopted the cheaper solution; the gap between what is rational and what is happening is the leader’s market position. If the crossover is still ahead but predictable, Wright’s Law gives you the trajectory and the timeline; a leader who builds the model before the crossing is not early, they are already in position when institutions finally move.

Before the crossing, every partnership comes with conditions, every funding relationship carries compromise risk, and institutional gatekeepers hold the leverage. After it, inertia runs the other way: the institution’s cheapest option is to adopt what the leader built, and the model is commercially viable by design, not as a side‑effect of mission. Hard problems stop looking like moral obligations and start looking like market positions.

Applied to AI displacement:

The legacy cost of institutional transition support – case management, in‑person training programmes, employment placement services – runs between US$8,000 and US$35,000 per worker per year in OECD systems. [3] That is the cost the state currently absorbs when a displaced worker enters extended unemployment.​ [3]

The current cost of AI‑assisted skills assessment, portable credential verification, and personalised reskilling pathway design is a fraction of that – and falling as platform competition increases and AI tooling becomes commodity infrastructure. Tools that once required institutional scale – labour‑market data, credential‑verification systems, skills‑matching algorithms – are increasingly available at near‑zero marginal cost on existing digital rails.​ [4]

The crossing point is not hypothetical. In most high‑income‑country displacement scenarios, a technology‑enabled transition model is already cheaper than 60 to 90 days of institutional unemployment support. After that crossing, the institution’s cheapest option is to fund the transition model rather than carry the cost of not doing so.

One move that brings the crossing point forward: portable credential verification on open digital‑identity rails, removing the proprietary lock‑in that currently keeps most reskilling outcomes platform‑dependent. A worker whose credentials are verified independently – not by the platform that trained them – has portable value that compounds across transitions. A worker whose credentials live inside a single platform’s ecosystem is a recurring‑revenue model for the platform, not a transition success.​ [5]

The commercial position at the Crossover Point: a transition model built on portable credential infrastructure, verified by independent income data, financed by the savings it produces in institutional unemployment support. That model becomes the rational choice for governments to fund not because it is mission‑aligned but because it is cheaper. That is a market position.

Incoming regulatory requirements – including AI‑Act‑style employment provisions and worker‑data‑portability frameworks under active consideration in the EU, UK, and Australia – would, if enacted, accelerate the crossing point further. The model does not depend on them. They are additional tailwind, not the engine.

2. The Chokepoint Map

What it is: A map of every point where an institutional actor can slow, veto, or tax the path from one Outcome Unit to scale. The critical distinction is between structural chokepoints – created by law, physics, or irreducible constraints – and legacy chokepoints, which persist because of habit, assumption, or incumbents who benefit from treating negotiable gates as permanent ones. In almost every hard‑problem domain, at least one chokepoint treated as fixed five years ago is now negotiable.

Applied to AI displacement:

People – access to displaced workers is a legacy chokepoint. Institutional employment services, unions, and sector‑specific bodies are the traditional routes, but not the only ones. Displaced white‑collar workers are reachable through professional networks, alumni associations, and peer communities that formed when institutional support proved inadequate. A leader with trusted proximity to those communities does not need institutional permission to reach the people the problem affects. Within this sits a harder sub‑dimension: the identity transition. Mid‑career professionals whose self‑concept is built around a role – paralegal, financial analyst, diagnostic coordinator – often need to adopt a skills‑based identity before they will engage with any programme. Trusted proximity means being able to reach workers at that moment, not just after it. Leaders who have navigated identity transitions in adjacent contexts – career pivots, sectoral shocks, organisational restructures – hold an asset no institutional programme replicates at the individual level.

Money – transition funding is shifting from structural to navigable. Government unemployment‑support funds exist, but routing through them attaches conditions that slow innovation. Outcome‑based financing – where funders are paid from the savings generated by successful transitions rather than for inputs – makes this chokepoint navigable for leaders whose Outcome Unit is precise enough to contract against. This is why the Outcome Unit must come first: imprecise outcomes cannot be outcome‑financed; precise ones can.

Data – credential verification and labour‑market intelligence is the structural chokepoint. The data that determines which skills are valued, which credentials are recognised, and which transitions endure is proprietary, held by platforms, staffing agencies, and credential bodies with strong incentives to keep it that way. This is not a legacy habit; it is actively defended. A serious transition model needs a data strategy that either routes around this chokepoint or builds open infrastructure that makes it irrelevant. [6]

That infrastructure already exists. W3C Verifiable Credentials and the 1EdTech Open Badges 3.0 standard allow credentials to be issued, held, and verified without routing through a proprietary platform. A credential built on these standards is controlled by the worker, verifiable by any employer with a compatible reader, and insulated from platform policy changes. The permissionless route is not a theory. It is a technical specification already in production use.

Legitimacy – institutional recognition of new credentials is a legacy chokepoint being defended on dissolving foundations. Most large employers still filter candidates through credential systems built for a labour market that no longer exists, even as the talent they need increasingly holds credentials those filters would exclude. The pull mechanism is talent scarcity, not institutional reform: in AI‑exposed occupations such as financial analysis, legal research, and diagnostic support, the pipeline of traditionally credentialled candidates is shrinking faster than demand. Employers who cling to legacy filters face compounding vacancies; those who update them gain early access to a growing pool of skilled, motivated, demonstrably capable workers. [7] Guild Education’s outcomes illustrate this: employers partnering with Guild to reskill existing workers see roughly 3.5‑times higher internal mobility and double the wage increases for participants, driven by retention and pipeline logic rather than altruism. [8] [9] The leader who builds transition models that produce verifiable, portable, skills‑based credentials is not waiting for institutional legitimacy; they are building the evidence base that makes it rational for institutions to update their filters.

The permissionless route: Trusted community access, plus outcome‑based financing, plus portable credential infrastructure on open digital rails produces a transition model that does not require institutional programme approval, proprietary platform partnerships, or legacy credential recognition to operate. That route exists now.

3. The Voltage Test

What it is: The test of whether a model still works when the founder is not in the room – specifically, when the government changes, the funder pivots, and the founding team moves on. Most programmes fail not because the idea is wrong but because the design was never required to answer this question. Chef‑dependency – models built around a founder’s presence, judgement, and relationships rather than replicable logic – is the structural cause. [9] Agentic AI is now a scaling architecture that lets you design for the Voltage Test from the outset, not patch for it at scale.

Applied to AI displacement:

Most reskilling programmes fail the Voltage Test the same way. They are built around the expertise of a founding team, dependent on government contract cycles or philanthropic funding, and tuned to the political moment in which they launched. When the contract ends or the moment shifts, the programme ends. Workers halfway through a transition are stranded. The credential infrastructure they were building remains proprietary to the programme that built it.

The chef‑dependent points in a workforce transition model are two: the trusted community relationships that give access to displaced workers at the moment of maximum vulnerability, and the employer relationships that make credential recognition credible. Both are solvable at the design stage. Community trust is buildable through peer facilitation rather than expert delivery; a model that trains recently transitioned workers as guides builds distributed trust infrastructure instead of relying on the founder’s relationships. Employer recognition is increasingly transferable as the evidence base for skills‑based hiring accumulates; credible evidence, not institutional mandate, is what changes behaviour.

Agentic AI shifts the calculation further. Skills assessment that once required a trained assessor can now be delivered through conversational tools at near‑zero marginal cost. Personalised reskilling pathway design that needed expensive human case management can be operationalised through AI that adapts to individual circumstances without founder oversight. A model that builds these tools into its architecture from the start is voltage‑proof in ways a model built around human expert delivery find harder to achieve.

Who is Making Ground

Organisations below are grouped by their relationship to this problem. The distinction matters: it shows what has been proven and what remains to be built.

Directly addressing this problem

Guild Education directly addressing workforce transition for incumbent workers - connects large employers to curated learning programmes, routing the employer’s existing education benefit spend toward business-aligned reskilling rather than ad hoc tuition reimbursement. In 2024, Guild enabled 60,000 programme completions across 1.4 million engaged employees, with participants recording wage increases 2 times larger than non-participants and internal career mobility 3.5 times higher. The employer pull mechanism is proven: retention and pipeline outcomes make the commercial case without requiring institutional mandate or philanthropic subsidy.

The structural limit is equally clear. Guild’s model is built inside the employer relationship - credentials are curated to the employer’s talent strategy, not portable to the open labour market. A worker whose Guild credentials are tied to their current employer’s internal mobility pathways is not holding portable value; they are holding employer-specific value. That distinction matters at the moment of displacement - which is precisely when portability is most needed. Guild demonstrates that employer-financed, outcomes-verified reskilling is commercially rational. It has not yet solved the portability problem that makes reskilling genuinely protective against displacement rather than merely useful for retention.

Structural lesson: Employer pull is real and financially quantifiable. The model that combines employer-financed outcomes with open credential infrastructure - not proprietary employer-specific pathways - is the architecture this problem needs and Guild has not yet built.

Opportunity@Work developed the STARS framework (Skilled Through Alternative Routes) to make skills-based credentials legible to employers who have historically required four-year degrees. It is working on the credential recognition chokepoint for displaced and non-traditional workers directly. The framework is designed to travel without Opportunity@Work present in every employer relationship - voltage architecture applied to the legitimacy chokepoint. Partial proof: demonstrates the chokepoint routing argument convincingly. The full Voltage Test at scale is not yet confirmed.

The hard power contrast is direct. Federal credentialling reform in the US - multiple executive orders and legislative pushes to expand skills-based hiring in the federal government - has produced limited adoption after a decade of effort. Opportunity@Work achieved measurable employer behaviour change by building the evidence base that made updating filters economically rational, rather than by mandating filter updates. Credible evidence, not compulsion, is what moves this chokepoint.

Structural lesson: The legitimacy chokepoint yields to evidence, not mandate. Soft power that builds open infrastructure changes employer behaviour faster than hard power that tries to legislate it.

Adjacent Domain, Structural Lesson Transfers

BRAC is not working on AI displacement. It has solved the voltage problem that every transition model faces. The graduation model - asset transfer, skills training, social protection, and access to financial services - is designed to run without BRAC present at every site. The financial architecture recycles participant contributions rather than depending on external grant renewal. Outcomes are verifiable by independent data rather than programme self-reporting. The structural logic transfers directly to a workforce transition context where the goal is verified income above baseline, not programme completion rates.

Structural lesson: Voltage-proof design is a founding decision, not a scaling retrofit. The model built from the outset to run without its founders is structurally different from one that adds replication mechanisms after the fact. Every major government workforce transition programme of the last two decades was built around a political moment rather than a durable architecture - and failed accordingly when the moment passed. BRAC’s model outlasts governments, funding cycles, and founding teams because it was designed to from the start.

Aravind Eye Care demonstrates the Crossover Point argument in a different hard problem domain. Aravind crossed the point at which providing eye care became cheaper than the system’s cost of preventable blindness - then built a model that delivers world-class outcomes at a fraction of the cost of comparable services, without philanthropic subsidy, without government mandate, and without its founding team present in every operating theatre.

Structural lesson: The Crossover Point is not a theory. It is an empirically demonstrated phenomenon across multiple hard problem domains. What follows it is a market position, not a mission. Government-funded eye care programmes in comparable settings have not achieved Aravind’s outcomes despite larger budgets - because the institutional incentive was programme delivery, not Outcome Unit achievement. Aravind’s independence from that incentive structure is what allowed it to optimise for the crossing point rather than the budget cycle.

The Gap

No organisation has yet assembled a fully voltage-proof, outcome-financed workforce transition model built on portable credential infrastructure, serving displaced white-collar workers in the specific occupational categories most affected by AI displacement, at the scale the problem requires. The structural lessons exist across the organisations above. The components exist. The architecture that combines them into a self-funding model - one that runs without institutional contract renewal, without founder presence, and without routing through the platforms that currently own the credential infrastructure - has not yet been built.

That gap is filled by soft power, not by budget or institutional mandate. Specifically: trusted proximity to displaced workers at the moment of maximum vulnerability, structural independence from the platform employment models capturing the transition, and the freedom to define success on the worker’s terms rather than the funder’s. Rather than being sector-specific, those characteristics are held by leaders who developed them in any context where independence from a more powerful actor determined what was possible.

The Market Opportunity

The window for building the right transition architecture is 18 to 36 months. Two consolidation trends are closing it at the same time. First, platform M&A: in December 2025, Coursera acquired Udemy in a US$2.5 billion deal, combining two of the world’s largest workforce‑credential ecosystems into a single platform serving 270 million learners. Second, LLM providers – OpenAI, Google, Microsoft – are building agentic career coaching directly into operating systems and productivity suites. When AI career guidance becomes ambient infrastructure rather than a standalone product, the interface through which displaced workers navigate transition will be owned by the same firms whose AI caused the displacement. Both trends narrow the window for independent credential infrastructure; 18 to 36 months is a live estimate, and the real window may be shorter. [10]

The default trajectory is already visible. Reskilling programmes multiply. Credentials proliferate on proprietary platforms. Workers complete programmes and discover their new credentials are legible only within the ecosystem of the platform that issued them. The transition succeeds by the platform’s definition of success – continued platform dependency – and fails by the worker’s, which is portable value and genuine income recovery. That trajectory requires no decision. It is what happens when the leaders who could build something different are still routing through the gatekeepers the permissionless route was designed to bypass.

The commercial position available to a leader who moves now is specific. A transition model built on the Outcome Unit defined above – portable credentials, independent income verification, 18‑month baseline – is contractable. It can be outcome‑financed against the savings it produces in institutional unemployment support. It does not require philanthropic subsidy. It does not require government programme approval. After the Crossover Point, the institution’s cheapest option is to fund it. That is not advocacy. That is a market position with a closing window.

The architecture that fills this gap does not require institutional permission to build. It requires proximity, independence, and a precise enough Outcome Unit to finance against. If you hold the first two, the third is where the conversation starts. A three‑way distinction helps: a leader with trusted proximity but no structural independence from platform employment models is a platform partner – their access serves the platform’s interests, not the worker’s. A leader with structural independence but no trusted proximity to displaced workers is a think tank – their analysis is credible but not deployable at the moment the problem is actually happening. A leader who holds both is the decisive architecture builder this brief is describing. That position is rare. It is not credential‑dependent, sector‑specific, or budget‑constrained. It is structural.

The window is the same for all of us. What we do with it is the decision this series is built around. The gap is mapped. What you bring to it is the next key question.

The Soft Power Assessment maps what you’re holding. If the result suggests this is your problem to move, that’s where I work - mapping soft power against hard problems like this one and finding where the leverage actually sits.

This problem is one of twelve. By Edition 12 you’ll have a clear picture of where your leverage is highest.

Leave a comment

Sources and Notes

[0] The problem: scale and trajectory: The IMF estimates roughly 40% of jobs globally face meaningful AI exposure, rising to approximately 60% in high‑income economies. The WEF Future of Jobs Report 2025 (building on the 2023–2025 cycle) projects 92 million roles displaced by 2030 alongside 170 million new ones created, a net gain of 78 million. The net figure masks a distributional problem: the jobs lost and the jobs created are not in the same sectors, geographies, or skill levels. Middle‑management job postings dropped more than 40% between April 2022 and October 2024 (Deloitte). US job openings in professional services hit their lowest level since 2013 in January 2025, a 20% year‑on‑year fall (BLS JOLTS). Approximately 40% of white‑collar job seekers in 2024 failed to secure a single interview (American Staffing Association). In low‑ and middle‑income countries, direct generative‑AI displacement risk is currently lower (ILO estimates 0.4% of jobs vs 5.5% in high‑income countries), but automation risk in manufacturing, logistics, and routine services affects a far larger share of the workforce. Sources: IMF, Gen‑AI: Artificial Intelligence and the Future of Work (2024); WEF, Future of Jobs Report 2025; Deloitte (2024); BLS (January 2025); American Staffing Association (2024); ILO, Generative AI and Jobs (2023).

[1] Crossover Point cost range: US$8,000-US$35,000 per participant per year reflects OECD public‑spending data on labour‑market programmes. OECD training spend for unemployed workers averaged 0.1% of GDP in 2023, a 30% fall since 2010. The 60-90 day crossover estimate is derived by amortising the US$8,000-US$35,000 annual OECD carry cost against the $500-$1,500 per-user licensing costs of enterprise reskilling SaaS, assuming a successful transition within one fiscal quarter. Sources: OECD/TUAC, Time to Activate Labour Market Policies (2025); OECD ALMP dataset; OECD Skills Outlook.

[2] Hard‑power workforce transition failures: US TAA lapsed July 2022, serving 40,000-60,000 workers annually in final years. The UK National Retraining Scheme, examined in NAO reporting on the adult skills system, was wound down by 2020 without reaching scale. Australia’s JobActive was redesigned four times 2003–2022; an ACOSS review found compliance was prioritised over outcomes. Sources: US Congressional Research Service; UK National Audit Office; Australian Centre for Social Policy / UNSW Social Policy Research Centre.

[3] Cost of institutional transition support: Same range as note [1]. US Trade Adjustment Assistance and the UK Flexible Support Fund sit toward the lower end; J‑PAL‑evaluated intensive programmes toward the upper end. Source: OECD Employment Outlook.​

[4] Declining cost of AI‑assisted skills infrastructure: AI assessment platforms (Workera, iMocha, HackerRank) now operate as enterprise SaaS with falling per‑assessment costs. The technical foundation for portable credentials is W3C Verifiable Credentials and Self‑Sovereign Identity - open standards allowing credentials to be issued, held, and verified without a proprietary intermediary. 1EdTech Open Badges 3.0 implements these standards and is widely adopted. Sources: platform documentation (2025); 1EdTech Digital Credentials Summit 2025; W3C VC Data Model 2.0 (2024).

[5] Portable credentials and platform lock‑in: Around 44% of Fortune 500 firms offer internal micro‑badge pathways tied to external providers. The 1EdTech Wellspring Initiative exists specifically to prevent credential lock‑in via open standards. Source: EdTech Breakthrough, Credentialing in the AI Economy (2025).

[6] The data chokepoint: Only about 15% of HR professionals can currently act on skills credentials in hiring decisions. Source: 1EdTech / HolonIQ Digital Credentials Summit 2025 reporting and HolonIQ 2025 Global Learning Outlook.

[7] Shifting employer credential preferences: 74% of employers in 2025 preferred verified digital skills credentials for AI‑related roles over degrees alone; job postings recognising verifiable AI certifications grew ~31% between 2024 and 2025. Sources: WEF Future of Jobs; LinkedIn Economic Graph / Workforce Report.

[8] Guild Education outcomes data: For 2024, Guild Education reports learners saw wage increases roughly 2× higher than non‑participants, were 3.5× more likely to move internally, with 60,000 programme completions and 1.4 million engaged. Revenue model depends on employer partnerships, not open worker‑owned credentials. Sources: Guild press release “Guild Has a Banner 2024” (Jan 2025); Sacra equity research on Guild (Feb 2026).

[9] Chef‑dependency and the Voltage Test: Draws on John List, The Voltage Effect (Currency, 2022) and University of Chicago Crime and Education Lab research on scale and ‘voltage drop’.

[10] The 18 to 36 month window: Strategic estimate, not a formal forecast, based on two consolidation trends: platform M&A (Coursera–Udemy, US$2.5bn, December 2025; 300+ EdTech deals in 2024) and LLM vertical integration (OpenAI, Google, Microsoft embedding career coaching into OS and productivity suites). Sources: ALM Corp Coursera–Udemy analysis; HolonIQ EdTech M&A data.

Since then, the story has shifted again. By mid-March 2026, Australia has moved beyond just helping to design the system; it is beginning to act as a de facto arbitrator of allied access to critical minerals, even as China still dominates much of the global midstream processing in key materials. That shift is visible in the federal government’s Critical Minerals Strategic Reserve, its updated critical minerals policy settings, and a more activist diplomatic agenda on critical minerals led by Minister Madeleine King’s recent visits to Canada and the US.

For leaders and investors, this matters because it turns abstract strategy into concrete institutions, contracts, and sites on the map.

Why critical minerals are now Australia’s leverage point

Critical minerals have moved from a niche topic to the centre of the US–China technology and security contest, as highlighted in work on the China challenge in critical minerals and recent coverage of a US‑led critical minerals trading bloc. They sit inside defence systems, electric vehicles, grid storage, advanced manufacturing and the digital infrastructure that underpins modern economies, a point underlined in the Australian Government’s Critical Minerals Strategy and prospectus and its updated critical minerals list.

Australia starts from three advantages:

• World‑class reserves of lithium, rare earths, nickel, cobalt and other critical minerals, documented in the government’s critical minerals prospectus.

• A reputation for regulatory reliability, environmental and labour standards, and alliance alignment with G7‑plus partners, reinforced in ministerial briefs on critical minerals cooperation and updated policy guidance.

• A web of bilateral and minilateral frameworks with the US, Japan, Korea, the EU and Canada focused on diversifying away from opaque, concentrated supply, including the October 2025 framework and subsequent King visit to strengthen critical minerals cooperation and US‑hosted critical minerals forums.

China still controls a dominant share of global mid‑stream processing for many of these minerals, but within emerging groupings such as ‘Pax Silica’ and G7‑plus forums, Australia is starting to shape how like‑minded partners secure access and set informal price bands. The question for 2026 has shifted beyond whether Australia will ‘stay a quarry’. Now, it is how far it is willing to go - legally, financially and politically - to lock in a role as a trusted arbitrator in a fractured system.

The AU$1.2 billion strategic reserve: financial architecture, not stockpiles

The first major shift since the original article is the move from high‑level frameworks to a fully funded Critical Minerals Strategic Reserve (CMSR).

The CMSR, foreshadowed in Treasury and Parliamentary Budget Office costings and detailed in government fact sheets and commentary such as this analysis of Australia’s critical minerals reserve architecture, allocates AU$1.2 billion in public capital to maximise the strategic value of Australia’s critical minerals. The design is deliberately different from Cold War‑style stockpiles:

• The reserve is slated to become operational from late 2026, with the government securing rights to production rather than stacking physical ore in warehouses, as outlined in official descriptions of the Critical Minerals Strategic Reserve and in explainer pieces like “Australia is betting on a new ‘strategic reserve’ to loosen China’s grip on critical minerals”.

• Initial focus is on gallium, antimony, and rare earths - materials that sit in semiconductors, radar, magnets, and other dual‑use technologies - rather than bulk tonnage, as reported by Reuters and referenced in government media statements.

Critically, the first actual transactions - likely securing rights to gallium and antimony - are contingent on finalising the Rare Earths Production Scheme and associated production‑linked support, which are still in the legislative and design phase. Think of this less as a warehouse and more as a financial architecture:

• In tight or weaponised markets, Canberra can redirect contracted volumes to allies and priority industries, a role spelt out in the PM&C overview of the reserve’s objectives.

• In normal times, the reserve’s presence shapes expectations about future availability, price floors and the terms on which ‘clean and clear’ supply will be offered, a point made in industry commentary like this Discovery Alert briefing on strategic resource security.

For boards and founders, this means critical minerals projects are no longer just exposed to spot prices and private offtake. They sit inside an emerging sovereign risk‑management tool that can stabilise revenue - or redirect it - when the system is under stress, as argued in analyses of strategic investment drivers in NT critical minerals and Australia’s critical mineral geopolitics.

Pricing the ‘green premium’: from ESG rhetoric to revenue support

In the first article, I argued that ESG had become the price floor for market entry, and that Australia’s brand could command a higher ceiling based on supply‑chain security. In 2026, that conversation is turning into actual price‑support mechanisms.

Two developments matter here:

• The government has signalled that the CMSR will use production‑linked revenue support, including Contracts‑for‑Difference (CfD)‑style tools that set a price ‘collar’ around eligible projects, an approach discussed in government releases on delivering Australia’s critical minerals supply and in industry briefs such as “Australia’s critical minerals reserve shifts from storage to strategy”.

• Industry groups such as the Association of Mining and Exploration Companies (AMEC) have pushed for explicit price floors to ensure Australian, ESG‑compliant producers are not wiped out when opaque, subsidised supply undercuts global prices, as outlined in the AMEC critical minerals strategic reserve framework and related policy commentary.

In practice, that looks like:

• A floor price that triggers support when global prices crash due to non‑market behaviour.

• A ceiling price that claws back taxpayer support when prices surge, sharing upside rather than socialising only the downside - mechanics that echo the logic of the Capacity Investment Scheme applied to critical minerals.

This is where Australia’s rule‑shaping push shows up internationally. In February and March 2026, Resources Minister Madeleine King has used visits to Washington and Toronto to make the case for coordinated approaches to price floors, market transparency and recognition of high ESG standards among the US, Canada, Japan, the EU and others, as detailed in her media release on talks with Canada and the US and in coverage of a proposed US‑led ‘Pax Silica’ grouping. Within that allied sphere, Australia is evolving from price‑taker to price‑shaper, even though it does not yet arbitrate the global market as a whole.

For investors and manufacturers, that signals the emergence of a de facto ‘green premium’ logic: compliant, traceable supply is more likely to benefit from stabilised revenue, long‑dated offtake and concessional finance than cheap, opaque alternatives.

Sovereignty by statute: Darwin and critical infrastructure law

Strategic infrastructure has become another arena where Australia is moving from signalling to enforcement.

The long‑running debate over the Chinese‑held lease of Darwin Port - and its intersection with defence posture and critical minerals export routes - has driven a broader rethink of how critical infrastructure is regulated. Analyses such as East Asia Forum’s piece on Australia’s Port of Darwin move and ISDS risks and The Diplomat’s take on why the leaseholder may welcome an Australian buyout illustrate the legal and diplomatic stakes.

In 2023–24, security reviews found ‘no current basis’ to overturn the 99‑year lease and the government opted for tighter oversight rather than rupture. By early 2026, that position has shifted. Canberra has now recommitted in principle to returning the port to Australian hands and is pursuing an active buy‑back - reported in the AU$600–900 million range - using updated Security of Critical Infrastructure (SOCI) Act powers as leverage rather than relying on blunt expropriation. Proposals like the Coalition’s plan to secure the Port of Darwin and evolving guidance from Infrastructure Australia show how quickly the policy consensus has hardened.

The signal to markets is that strategic infrastructure linked to defence posture and critical mineral export routes will be brought back inside a clearer sovereignty and national‑interest frame - even if that requires complex commercial and legal choreography rather than a headline‑driven decree. For global investors, that raises the bar on political‑risk analysis - but it also clarifies the rules. Future deals will be assessed not just on price, but on how they fit into an emerging national‑interest lens around critical minerals.

Middle Arm: building the ‘trusted hub’ for shipping chemicals, not ore

If the strategic reserve is the financial engine of Australia’s new role, the Middle Arm Sustainable Development Precinct in Darwin is its most visible physical manifestation.

The Northern Territory Government’s vision for Middle Arm is laid out in project documents such as this Middle Arm Sustainable Development Precinct overview and Infrastructure Australia’s common‑user infrastructure brief. The precinct is intended to support a multi‑user industrial hub that underpins a AU$40 billion economy by:

• Hosting LNG exports and gas‑related industry.

• Providing common‑user infrastructure for green hydrogen, green ammonia and critical minerals processing.

Company announcements and government documents describe Middle Arm as ‘purpose‑built for critical minerals processing and clean energy production’, with proponents planning to refine copper, cobalt and other critical minerals on site rather than shipping concentrates offshore, as summarised in local MP briefings such as this Middle Arm FAQ.

The reality on the ground is more complex:

• FOI releases and environmental groups highlight delays, emissions concerns and contested social licence, particularly around impacts on Darwin Harbour and local communities, as reported by the ABC in coverage of Middle Arm delays and detailed in the Environment Centre NT’s Middle Arm factsheet.

• The strategic environmental assessment is ongoing, and final investment decisions on common‑user infrastructure have been pushed back under the weight of FOI‑driven scrutiny and litigation, even as the Commonwealth has flagged up to AU$1.5 billion in equity support.

• Proponents and the NT government emphasise jobs, regional development and the chance to anchor downstream processing - and therefore more value capture - onshore, points stressed in official Middle Arm project materials and government FAQs.

For leaders watching from elsewhere, Middle Arm is a live test of whether Australia can align three things at once: downstream value capture, robust social and environmental standards, and credible engagement with Traditional Owners and local communities. Without that alignment, the precinct risks becoming a symbol of contested transition rather than trusted arbitrator.

What this adds up to for decision‑makers

Taken together, the 2026 moves turn Australia’s critical minerals story from aspiration into architecture:

• A funded Critical Minerals Strategic Reserve that will sit at the centre of how allied partners access key inputs when markets are tight, once the first REPS‑linked transactions are in place..

• Emerging price‑support instruments and policy settings that operationalise a green premium and shield compliant producers from the worst of non‑market volatility, initially within a Pax Silica / G7‑plus sphere rather than the entire global market.

• Tougher critical‑infrastructure and foreign‑investment rules, signalled in debates over Darwin Port and strategic precincts like Middle Arm, that give Canberra more options to re‑align assets with national interest and allied strategy.

• A contested but pivotal Middle Arm precinct that could physically anchor the shift from exporting ore to exporting refined products into trusted value chains - if social licence, environmental safeguards and commercial timelines can be brought into alignment.

If the 2025 story was about Australia learning to use its soft power and hard assets to become a ‘strategic architect’, the 2026 story is about a middle‑sized economy testing how far it can go in arbitrating flows and prices without over‑reaching.

For boards, founders and policymakers, three practical questions flow from this:

1. How exposed is your strategy to the emerging architecture of reserves, price floors and ESG‑linked access?

2. Are you treating Australian law, infrastructure designations, and social licence as constraints - or as levers you can align with?

3. If you operate in another resource‑rich jurisdiction, what would your own version of a strategic reserve, price‑support collar and ‘trusted hub’ precinct look like?

Those are the questions I am working on with leaders who want to play a bigger game in a more fractured system. If you’d like to explore how this applies to your sector or geography, you can read the original article below, or reach out directly.

Leave a comment

The problems this series examines affect hundreds of millions or billions of people, cross borders, and sit inside systems - laws, markets, infrastructure, and norms - that no single institution controls and no single actor can fix alone. That is what we mean by hard.

The cost assumptions were formed, for example, when satellite imagery required government contracts and physical surveys cost hundreds of dollars per household. The timeline expectations were formed when reaching a billion people required a decade of institutional negotiation. The dependency assumptions were formed when the only route to scale was via multilateral agencies, bilateral funders, and the slow physics of government procurement cycles.

Those assumptions were reasonable once. They are now actively misleading.

This series introduces three calculations that turn hard problem rhetoric into concrete numbers; numbers that de‑risk working on hard problems and clarify where to put scarce time, capital, and attention.

The Conditions Have Changed

The tools available to a leader working on a hard problem today are not incrementally better than the tools available in 2010. They are categorically different in cost, reach, and speed.

Across sectors, the unit economics of key enabling technologies have collapsed:

• Global average costs for digital identity verification checks are now around US$0.20 and are projected to fall further as volumes scale.

• High‑resolution satellite imagery and remote sensing have driven the per‑parcel cost of land measurement far below that of legacy, labour‑intensive field surveys, making parcel‑level mapping technically and financially feasible at national scale.

• India’s Unified Payments Interface (UPI) processed about 172 billion transactions in 2024 - over 14 billion per month - illustrating the scale of low‑cost, real‑time digital payments.​

• Brazil’s instant payment system Pix has reached more than 170 million users, roughly 93 percent of the country’s adult population, without relying on dense physical branch networks.

• AI‑powered tools for legal and compliance workflows are already delivering large time and cost savings, suggesting that specialised land‑tenure documentation can similarly see order‑of‑magnitude reductions in per‑case costs. For example, AI document intelligence platforms automate review, extraction, and drafting across complex legal documents.​

• In agriculture, emerging AI agronomy platforms use data‑driven recommendations to scale soil analysis, crop planning, and pest detection across many farms, sharply lowering the marginal cost of advice once the system is deployed.​

These demonstrate changed conditions. The hard problems are still hard. The people still waiting for socioeconomic, environmental, health, and education solutions are still waiting. What has changed is the set of tools available to the leaders who decide to act - and therefore what we should expect of those leaders, and of ourselves.

The change in conditions is about what existing leaders can do. And it is about who can lead.

For most of the last half-century, working seriously on hard problems required a specific kind of entry ticket: institutional affiliation, large funding bases, decades of relationship-building with the gatekeepers who controlled access to beneficiaries, governments, and legitimacy. The problems were large enough that only large actors could reach them. Which meant the universe of people who could credibly act was, by definition, small.

That constraint is dissolving. The same cost curves that have made high‑resolution satellite and GPS-based mapping vastly cheaper than traditional household field surveys now make them accessible to a team of twenty as readily as to a multilateral agency. The same digital identity rails that have pushed average per‑person onboarding costs down to well under a dollar do not discriminate between an established NGO and a leader who incorporated last year. The infrastructure that used to require institutional scale to access is becoming, in domain after domain, a public good. Which means the question of who is structurally positioned to move a hard problem is no longer answered by looking at who has the biggest budget or the longest track record. It is answered by looking at who has run the three calculations - and who is willing to act on what they find.

There is a third implication that follows from all of this: some leaders who run the three calculations will find that the structural agency available to them has already outrun the ambition they’ve been willing to declare. That the limiting factor is no longer infrastructure, capital, or institutional permission. That what they’ve been calling ‘being realistic’ is just staying calibrated to an older world.

The three calculations have a side effect. On the one hand, they make bolder goals possible. On the other hand, they make smaller goals harder to justify.

Not more ambition. More precision. Specifically: three calculations that the most effective organisations working on hard problems are already running, and that most leaders have not yet made.

This series applies the three calculations to twelve of the world’s hardest problems. This piece explains what the calculations are, where they come from, and why all three are necessary before any of them is sufficient.

Subscribe now

The First Calculation: The Crossover Point

In the 1930s, Theodore Wright was studying the economics of aircraft manufacturing when he noticed something that would later be generalised across almost every industry in the world. Every time cumulative production of a given technology doubled, the cost fell by a predictable percentage. Not randomly. Not occasionally. Reliably, across industries, across decades, across technologies with nothing in common except that humans were making more of them and getting better at it.

Wright’s Law is now one of the most empirically robust findings in the economics of technology and cost decline, as shown in this overview of Wright’s Law and learning curves. It has been validated in solar panels, semiconductors, batteries, DNA sequencing, and dozens of other technologies. The implication for hard problems is specific and underused: the cost of the tools needed to address most hard problems is falling on a predictable curve. Which means the moment at which a structural solution becomes cheaper than the system’s cost of doing nothing is calculable - not guessed at, not hoped for, but calculated.

That moment is the Crossover Point.

The Crossover Point asks one question: when does a structural solution become cheaper than the system’s cost of doing nothing - and what will it take to bring your model to that point faster?

Before it, a leader working on a hard problem is competing against inertia with more expensive tools than the incumbent. Every partnership comes with conditions. Every funding relationship carries compromise risk. The institutional gatekeepers hold the leverage because they control the only viable route to scale.

After it, inertia works in the other direction. The institution’s cheapest option is now to adopt what the leader built - because it’s cheaper than continuing to do nothing. This is also when the model becomes commercially viable - not as a byproduct of mission, but structurally. After the Crossover Point, a leader is not competing against a better-funded incumbent. They are operating with a lower cost base than the institution whose inertia they just outran.

The cost of inaction is not zero. It is large and measurable. We can see it in three domains that are unusually well quantified:

• Every US$1 invested in deworming returns an estimated US$169 in productivity and long-term earnings.

• Rigorous analyses suggest every US$1 invested in family planning can generate on the order of US$7 in direct public-sector savings in high-income settings, and as much as US$100+ in wider social and economic benefits in low- and middle-income countries.

• Best-available analyses suggest that investing in agricultural research and development can return more than US$20–30 in benefits for every US$1 invested.

We can calculate the moment, yet the economics for a solution do not stop there. The calculation has a trajectory, not just a moment. Solution costs fall on a predictable curve as technologies scale. At the same time, the cost of institutional inaction compounds - more displacement absorbed, more productivity lost, more systemic fragility carried forward. Both curves are moving. Which means the crossing point arrives sooner than a static comparison suggests.

The cost of institutional inaction - in lost productivity, compounding health burden, and systemic fragility - is calculable in almost every hard problem domain.

Hard problems aren't waiting for more resources or more political will. They're waiting for leaders running a different set of calculations. When we run them, hard problems stop looking like moral obligations and start looking like market positions.

A leader who builds the model before the crossing occurs is not early. They are positioned. When institutions finally move - and they will, because inaction becomes more expensive than adoption - they will find the architecture already built by someone else.

In almost every domain, the Crossover Point exists, is calculable, and yet is rarely calculated first. It should be.

The Second Calculation: The Chokepoint Map

For decades, institutional intermediaries controlled the only viable path to scale: to beneficiaries, to funders, to governments, and to legitimacy. That control is eroding - in some places, it has already gone.

Digital public infrastructure shows what this looks like in practice:

• India’s Unified Payments Interface (UPI) and Brazil’s Pix have made low‑cost, real‑time payments available at population scale on rails that no single institution owns.

• Digital identity and eKYC have pushed per‑person onboarding costs down to well under a dollar in many markets.​

• Community members in countries like Ghana are using mobile and geospatial tools to help map land boundaries and feed data into formal systems.​

• AI‑assisted legal tools are expanding what paralegals and front‑line legal workers can do by automating routine tasks and freeing scarce central expertise for the edge cases.

Rather than being incremental improvements, they demonstrate what happens when essential infrastructure becomes a public good rather than a proprietary asset.

Every chokepoint a leader is still routing through is a point of leverage someone else holds over their mission. The organisations working most effectively on hard problems are the ones that mapped those chokepoints early - and either routed around them or built the infrastructure that made them irrelevant. That is not just operational efficiency. It is the recovery of agency. It is how an organisation earns the structural ability to say no to a bad partnership, decline a distorting grant, or hold a position under pressure without fearing the consequences.

The Chokepoint Map has four domains: People, Money, Data, and Legitimacy.

In each domain, the question is the same: who can still slow, veto, or tax your path from Outcome Unit* to scale? (*explained below)

The critical distinction within each domain is between structural chokepoints - those that exist because of law, physics, or irreducible constraint - and legacy chokepoints - those that exist because of habit, assumption, and the interests of actors who benefit from treating negotiable gates as permanent ones.

In almost every hard problem domain, at least one chokepoint treated as permanent five years ago is now negotiable. Finding it before incumbents consolidate around the new tools is a strategic advantage with a closing window. The Chokepoint Map is how you find it - and how you distinguish the gates worth routing around from the ones that are genuinely structural.

The Third Calculation: The Voltage Test

Economist John List identified what he called the voltage drop in his book The Voltage Effect: How to Make Good Ideas Great and Great Ideas Scale: the phenomenon by which a large share - often 50 to 90 percent - of programs that work in pilots fail at scale. This was not caused by bad ideas or insufficient funding. It was chef-dependency - models built around a founder’s presence, judgment, and relationships rather than replicable logic. When the founder moved on, the funding changed, or the political context shifted, the model lost its voltage. In addition to it stopping to work, it also started making compromising decisions on the way down - accepting distorting funding, narrowing its mandate, or drifting from its original purpose - because it had no structural alternative.

The Voltage Test asks one question: does your model work when you’re not in the room?

More specifically: does it work when the government changes, the funder pivots, and the founding team moves on? Most programs fail this test not because the idea is wrong but because the design was never asked to answer it. The founder’s presence was assumed. The funder’s continuity was assumed. The political moment was assumed. When those assumptions broke, the model broke with them.

The organisations that score highest in any serious assessment of hard problem impact have all solved this problem - though not all the same way.

• Pratham’s Teaching at the Right Level works without its founders in the room and has been adapted and implemented by governments and partners across India, Africa, and beyond, reaching more than 80 million children.

• Namati built the Grassroots Justice Network - now over 16,600 members from 175 countries - on a model designed from the outset to put the power of law in people’s hands rather than central offices.

• Landesa works with governments to design and implement national land laws and programs that are field-tested, financially and politically feasible, and explicitly intended to function at scale without Landesa present in every community.

• One Acre Fund uses a revolving loan structure in which farmer repayments are recycled into the following season, enabling the model to operate and grow through system design rather than founder oversight.

What made building voltage-proof models hard was time. Historically it took a decade of iteration to embed a model’s logic in systems rather than people. That constraint is dissolving. Agentic AI - systems that execute multi-step processes, adapt to local variation, and operate without constant human oversight - is more than a productivity tool. It is a scaling architecture. The Voltage Test is now answerable in the design phase, not the scaling phase. Leaders who ask ‘how does this work without me?’ before they ask ‘how do we grow?’ are building something structurally different from every previous generation of mission-driven organisation.

Why All Three Are Necessary

Each calculation is insufficient alone.

A leader who calculates the Crossover Point but hasn’t mapped the chokepoints will arrive at the right moment through the wrong route - dependent on institutional gatekeepers who will extract rent for access at exactly the moment the leader needs independence most.

A leader who maps the chokepoints but hasn’t run the Voltage Test will build something that works brilliantly in the pilot and collapses when they leave - because the model’s logic lived in their judgment rather than in replicable systems.

A leader who passes the Voltage Test but hasn’t calculated the Crossover Point will build something durable and permanently subscale - because the economic logic that would make it the rational choice for governments and commercial capital to adopt never got calculated, and the model stays dependent on philanthropic subsidy when it could have been self-funding.

Together, the three calculations produce something different from any one of them: a map of leverage. Not analysis of the problem - every domain has that. A map of when to move, how to move without permission, and what to build so that what you build holds.

The Map The Three Calculations Produce

The three calculations are not a checklist. They are three different lenses - economic, structural, and architectural - that produce different information. Their value compounds at the intersections.

We can picture them as three overlapping circles:

1. The Crossover Point and the Chokepoint Map together - without the Voltage Test - produce the Opportunity Window. You know when the economics work and you know the routes that don’t require institutional permission. You can see that a move is possible and viable. What you cannot yet see is whether what you build will hold when you step back. This is where most well-funded initiatives live: they found the right moment and got in through the right door, but the model depends on the founder’s continued presence to function. The window is real. The model won’t last.

2. The Chokepoint Map and the Voltage Test together - without the Crossover Point - produce the Durable but Subscale Model. You know how to move without institutional permission and you’ve built something that holds without you. The routes are clear and the architecture is sound. What you haven’t calculated is when the economics tip in your favour - so the model stays dependent on philanthropic subsidy rather than becoming the rational choice for governments and commercial capital to adopt. This is where many of the best mission-driven organisations live: genuinely independent, genuinely voltage-proof, permanently undercapitalised because the commercial case was never made.

3. The Crossover Point and the Voltage Test together - without the Chokepoint Map - produce the Stranded Asset. You know the economics work and you’ve built something durable. But you’re still routing through institutional gatekeepers to reach the people the problem affects. The model is real and the timing is right, but the route is controlled by someone else. This is where promising models get captured - absorbed by the institutions they were meant to route around, or forced into compromise partnerships because the permissionless alternatives were never mapped.

4. At the centre - all three calculations together - is the map of leverage itself. When to move. How to move without permission. What to build so that what you build holds. This is also where commercial viability and mission integrity stop being in tension. The Crossover Point tells you when the economics work. The Chokepoint Map tells you how to capture that value without routing it through extractive intermediaries. The Voltage Test tells you how to build something that compounds rather than depletes. Together they describe an organisation that doesn’t have to choose between surviving and doing what it said it would do.

Across twelve editions, this series will show where current real-world responses to each problem are stuck - which intersection they’re in, and what the missing calculation would change. The pattern that emerges is the argument.

The Outcome Unit

Beneath all three calculations is one prior requirement that most leaders working on hard problems have not met: the Outcome Unit.

The Outcome Unit is the atomic, binary, verifiable result:

• Not ‘education’ - one child reading independently at the age they should be.

• Not ‘land rights’ - one household with verified, digitally registered tenure.

• Not ‘financial inclusion’ - one person with an active account used in the last 90 days.

• Not ‘health access’ - one person who received the treatment the evidence says works, verified by independent data.

Without a precise Outcome Unit, the Crossover Point cannot be calculated - you cannot price what you cannot define. The Chokepoint Map cannot be completed - you cannot identify the route to the outcome if the outcome is a category rather than a result. The Voltage Test cannot be passed - you cannot embed replicable logic in systems if the logic depends on a founder’s judgment about what counts as success.

The Outcome Unit is not academic precision for its own sake. It is the foundation that makes agency real. A leader can only credibly say no - to a distorting grant, a compromising partnership, a mandate that would narrow the mission - if they can prove they are winning without the partner they are declining. The Outcome Unit is what makes that proof possible.

Share

This Series

The Soft Power Brief is a twelve-part series on the problems the world is still waiting to solve. Each edition takes one hard problem from current expert analysis and runs all three calculations on it - showing what the Crossover Point looks like in that domain, where the legacy chokepoints are and what has made them negotiable, and what a voltage-proof model would need to look like to hold at scale. The framework in this series comes directly from work I do with leaders and organisations who are turning their soft power into leverage on hard problems.

The problems are different. The calculations are the same. Across twelve domains, the same pattern appears: institutions fail, technology companies move into the space they left and monetise the problem rather than solve it, and a specific kind of leader fills the gap between the two. That leader is not defined by sector, credential, or budget. They are defined by structure. And that structure is identifiable before they act - not just recognised after they succeed.

The barriers to serious work on hard problems are falling. The leverage available to a leader who runs the three calculations is rising. What remains - the only thing that remains - is the question of which problem you care enough to drive a solution to, and whether your strategy is calibrated to the conditions that actually exist right now.

That is a personal question before it is a strategic one. The Soft Power Assessment at softpowerindex.lovable.app will show you what you are actually holding across six domains - not credentials, not track record, but the structural assets that determine where your leverage is highest. Twelve hard problems follow. By the end, you will have enough to decide where your impact will be highest over the next one to five years.

By Edition 12, if you’ve read and played along, you’ll know three things:

1. which problems your soft power fits,

2. which ones you should stop pretending you’re the right person to lead, if any, and

3. if you've been wondering whether you have a role to play - which problem you'll commit to over the next one to five years.

We’ve been operating inside a frame someone else set. This series is the way out of it. What’s on the other side is a clearer picture of what we are all capable of - and what becomes possible when we each act on that.

I look forward to embarking on this together.

Leave a comment

Why this? Why me? Why now?

They began as investor questions - the three tests a founder must pass before capital commits. They have since become important leadership questions of the times - for executives, institutional leaders, funders, advisors, and anyone working on a problem that matters at scale.

Each question probes something specific.

• Why this tests whether the problem is real, worth solving, and matched to your particular position.

• Why me tests whether you hold the structural characteristics that create genuine comparative advantage - not credentials, but proximity, independence, and freedom to act.

• Why now tests whether the moment is calculable, not just felt.

The conditions for answering all three have changed. The problems haven’t changed - the people still waiting are still waiting. Rather, the tools, the infrastructure, and the entry requirements for serious work on hard problems are different from what they were a decade ago. Which means the answers available to a much wider group of leaders are different too.

That is what this piece is about.

What is a Hard Problem - And Why Hard Power Can’t Solve It

Hard problems are not difficult projects. They are not complex organisational challenges or ambitious strategic goals. They are a specific category of problem with specific structural features that make them resistant to the tools most leaders have been trained to use.

A hard problem affects hundreds of millions of people across borders. It sits inside systems - laws, markets, infrastructure, norms - that no single institution controls and no single actor can fix alone. The cost of doing nothing compounds every year. And the problem persists not because nobody cares, but because the actors with the most power to address it have structural reasons not to - the incentives, the mandates, and the political economies all point away from the solution.

That last feature matters. Hard problems are not waiting for more resources or more political will. They are waiting for a different kind of power.

The world still rewards hard power. Budget size. Institutional mandate. The ability to legislate, compel, and sanction. These are the tools that built most of the systems we operate inside - and they are the tools that most leaders are still measured against. A government that passes a law, a corporation that deploys capital at scale, a multilateral agency that coordinates a global response - these are still the default answers to the question of who solves big problems.

But hard power has a specific failure mode on hard problems. It arrives late, with blunt instruments, after the political pain has become undeniable - and by then the window for the most effective intervention has often closed. It requires permission from the institutions whose interests are often entangled with the problem. It produces solutions that work when the political moment holds and collapse when it shifts. And it cannot cross the borders, sectors, and systems that hard problems operate across - because hard power is jurisdictional by design.

Soft power is different in kind, not just in degree. In this context, soft power is not charm, reputation, or brand. It is the structural ability to get outcomes without needing anyone else’s permission. It shows up as the freedom to decline misaligned funding, resist pressure to dilute a model, convene actors across lines that institutional mandates cannot cross, and hold a position under pressure without fearing the consequences - because the model’s viability doesn’t depend on the relationship being protected.

Soft power is the right tool for hard problems for a specific structural reason: hard problems require coordination across actors who cannot compel each other. No single government can legislate a billion people into literacy. No single company can mandate food security. No single institution can enforce the legal empowerment of five billion people. What moves these problems is not compulsion - it is models that others want to adopt because they are cheaper, easier, and more durable than the status quo. That is exactly what soft power, built deliberately, produces.

The leaders who are moving hard problems today are not the ones with the largest budgets or the strongest mandates. They are the ones who understood early that the tools of hard power were the wrong instrument for this category of problem - and built something different instead. The Soft Power Index tracks and ranks the organisations doing exactly that - ranked not by size or spend, but by the structural characteristics that make their models durable, independent, and replicable at scale.

Subscribe now

The Conditions That Have Changed

For most of the last half-century, the answers to these three questions were constrained by physics. “Why me?” was answered by institutional affiliation and budget size - because those were the prerequisites for reaching problems at scale. “Why now?” was answered by funding cycles and political windows - because those were the only engines powerful enough to move large problems. “Why this?” was answered by proximity to institutional mandates - because that is where the resources were.

Those constraints are dissolving. The tools available to a leader working on a hard problem today are categorically different from those available a decade ago. Digital identity verification dropped from US$23 per person to US$0.50. Satellite imagery with parcel-level resolution costs cents per hectare. AI-assisted legal documentation reduced costs from US$50-200 per household to US$5-15. AI agronomic advice - soil analysis, weather-optimised planting calendars, pest detection - is deliverable through an existing field officer network at near-zero marginal cost.

They are arguments about changed conditions. The hard problems are still hard. The people still waiting are still waiting. What has changed is the set of tools available to the leaders who decide to act - and therefore what we should expect of those leaders, and of ourselves.

The change in conditions are about what existing leaders can do. And, it is about who can lead.

For most of the last half-century, working seriously on hard problems required a specific kind of entry ticket: institutional affiliation, large funding bases, decades of relationship-building with the gatekeepers who controlled access to beneficiaries, governments, and legitimacy. The problems were large enough that only large actors could reach them. Which meant the universe of people who could credibly act was, by definition, small.

That constraint is diminishing. The same cost curves that make satellite imagery accessible at cents per hectare make it accessible to a team of twenty as readily as to a multilateral agency. The same digital identity rails that brought onboarding costs from US$23 to US$0.50 do not discriminate between an established NGO and a leader who incorporated last year. The infrastructure that used to require institutional scale to access is becoming, in domain after domain, a public good. Which means the question of who is structurally positioned to move a hard problem is no longer answered by looking at who has the biggest budget or the longest track record. It is answered by looking at who has run the three questions honestly and who is willing to act on what they find.

There is a third implication that follows. Some leaders who answer these questions honestly will find that the structural agency available to them has already outrun the ambition they’ve been willing to declare. That the limiting factor is no longer infrastructure, capital, or institutional permission. That what they’ve been calling “being realistic” is just staying calibrated to an older world.

Answering these questions has a side effect. It doesn’t just make bolder goals possible. It makes smaller goals harder to justify.

Why This?

The problems worth answering this question about are the ones that affect hundreds of millions of people across borders - hunger, illiteracy, displacement, the absence of legal identity, the lack of basic healthcare. Problems that sit inside systems the world runs on: laws, markets, infrastructure, norms. Problems where the cost of doing nothing compounds every year, and where no single government or company has either the incentive or the reach to solve them alone.

Answering “why this?” well requires two things many leaders skip.

The first is a precise Outcome Unit - the atomic, binary, verifiable result that defines what solving the problem actually means. Not “education” - one child reading at grade level. Not “land rights” - one household with verified, documented tenure. Not “financial inclusion” - one person with an active account used in the last 90 days. Not “health access” - one person who received the treatment the evidence says works, verified by independent data.

Without a precise Outcome Unit you cannot calculate the cost of inaction, you cannot map the route to scale, and you cannot prove you are winning without the partners whose conditions would compromise your mission. The Outcome Unit is not operational detail. It is the foundation that makes agency real.

The second is an honest assessment of whether this problem fits your specific position - not your interest, not your values, but your structural proximity to what is actually failing. The leaders who move hard problems are not always the ones who care most. They are the ones who understand most precisely what is broken and why - because they have been close enough to see it in a way that others haven’t.

Clarity on “why this?” turns concern into strategy.

Why Me?

This is the question most leaders answer with credentials. Sector experience. Track record. Organisational mandate. Those things matter - but they are not what creates genuine comparative advantage on a hard problem.

What creates genuine comparative advantage is structural position. Specifically: three things that cannot be bought, manufactured, or conferred by an institution.

Proximity to what is actually failing - not proximity to the analysis of what is failing, but to the specific moment when the problem becomes real for the person it affects. The leader who has sat with a community while a carbon developer walked them through a contract that would trade their land sovereignty for a one-time payment understands something about that problem that no policy brief captures. The domain doesn’t matter. The proximity does.

Independence from the institutional interests that created the problem - the ability to name publicly what is in the way, including the actors with the most resources to defend the status quo, without losing access to the rooms where decisions are made. Most people with institutional standing self-censor to protect it. Structural independence from that pressure is rare. Where it exists, it is among the most valuable positions a leader working on a hard problem can hold.

Freedom to define success on the problem’s terms rather than the funder’s - which requires the Outcome Unit to be precise enough that you can prove you are winning without anyone whose conditions would compromise the definition.

These are soft power assets. They are not sector-specific. A leader who developed them in climate finance holds them just as credibly as one who developed them in global health. The domain you built them in doesn’t determine where you can deploy them. The Soft Power Assessment at will show you where you stand - not as a credential check but as a structural map of where your leverage is highest.

The clearer your answer to “why me?”, the clearer your comparative advantage.

Share

Why Now?

This is the question most leaders answer with intuition. It feels like the right moment. The political window is open. The technology is ready. There is funding available.

Those are often observations. “Why now?” answered well is a calculation.

The Crossover Point is the specific moment at which a structural solution to a hard problem becomes cheaper than the system’s cost of continuing to do nothing. It is grounded in Wright’s Law - the empirically robust finding that costs fall predictably as technologies scale. Which means the moment at which acting becomes the rational choice - not the heroic one, not the idealistic one, but the economically rational one - is calculable in advance.

Before the Crossover Point, you are competing against inertia with more expensive tools than the incumbent. Every partnership comes with conditions. Every funding relationship carries compromise risk. After it, inertia works in the other direction - the institution’s cheapest option is now to adopt what you built. That is also the moment the model becomes commercially viable - structurally. After the Crossover Point, you are not competing against a better-funded incumbent. You are operating with a lower cost base than the institution whose inertia you just outran. That is not a mission. That is a market position.

The cost of inaction is not zero. It is large and measurable. Every US$1 invested in deworming returns an estimated US$169 in productivity and long-term earnings. Family planning programmes return US$60 to US$100 per dollar spent. The cost that institutions absorb from doing nothing - in lost productivity, compounding health burden, systemic fragility - dwarfs the cost of solving the problem. The Crossover Point is the moment those two cost lines cross.

Answering “why now?” well means calculating that crossing point - and understanding what would bring it forward. The sharper your sense of why now, the stronger your case for acting before someone else does - and before the window that currently exists closes around a different set of actors.

The Three Questions Together

“Why this?” gives you the problem and the Outcome Unit. “Why me?” gives you your structural position and where your leverage is highest. “Why now?” gives you the timing and the commercial logic for acting before the window closes.

Together they are the personal diagnostic that precedes rigorous analysis - the questions you answer to yourself before you run the numbers. They do not replace strategic precision. They ensure that precision is pointed at the right problem, by the right person, at the right moment.

Leaders who can answer all three clearly - not to investors, not to boards, but to themselves in the first instance - are the ones who build things that hold. They act from structural clarity rather than institutional permission. They define success on the problem’s terms rather than the funder’s. They know when to move because they calculated the crossing point rather than waited for consensus.

The Soft Power Brief is a twelve-part series on the problems the world is still waiting to solve. Published here on Substack from mid-March, it applies this same logic to twelve hard problems, one by one. Each edition runs three calculations on a specific problem and maps the leverage available to a leader who is structurally positioned to move it. By the end of the series, my hope is that we will each have enough to answer all three questions for the problem that is ours to move.

The problems the world is still waiting to solve are not waiting for better institutions or bigger budgets - even though they often feels true. They are waiting for leaders who have answered three questions thoroughly - and are ready to act on what they found.

Why this? Why me? Why now?

A fourth question - added by a reader

Since this piece was published, raised a question that belongs in the framework:

With Whom?

It is the right addition. The three questions establish your own position - the problem, the fit, the timing. But soft power is relational by nature. The structural ability to convene across lines that institutional mandates cannot cross is not just a personal asset - it is how leverage compounds.

What soft power enables specifically is identifying partners who hold what you don’t. Different proximity to the problem. Different independence from different institutional interests. Different freedom to name different things publicly. Two leaders with identical positions add presence. Two leaders with structurally different positions add reach.

The conditions that have lowered the barriers to serious work on hard problems have also made it possible to find, reach, and work with those people - across sectors and borders that previously made that kind of partnership slow and rare.

A fourth question worth adding to the list: With whom - and what does their soft power reach that mine doesn’t?

Leave a comment

Sixty-six years later, Shri Mahila Griha Udyog Lijjat Papad is a 45,000-member worker cooperative with ₹1,600 crore (~US$192M) in annual turnover, exporting products to 25+ countries. Governments from Uganda and beyond have sent delegations to study what they have done.

One decision drove that success. They decided to accept no donations and no external investment. Every member is a co-owner with an equal share in profit and loss, and that structure has held through every financial crisis and growth challenge in the 66 years since.

This is Soft Power solving a Hard Problem. It is also a lesson in how to scale a solution without external capital, political backing, or a famous founder.

The Hard Problem

A Hard Problem is one that affects millions or billions of people. It can be stated in one sentence, and it sits in a gap that neither markets nor governments will fill - markets because it is not profitable enough, governments because it requires commitments that cross borders and decades.

Lijjat’s Hard Problem is women’s economic exclusion. Can a worker-owned business, with no hierarchy, no external capital, and no male ownership, lift women out of poverty at scale and compete with multinationals while doing it?

Yes, and they have been doing it for six decades.

The founding decision made that possible. By refusing donations from day one - including during financial crises - the cooperative’s credibility comes entirely from its own results. Every member owns the outcome. Quality and accountability are not imposed from outside. The members are the business model.

Subscribe now

The Soft Power

Soft Power is the ability to attract others to your model rather than forcing or paying them to follow it. It operates through credibility - when what you have built is specific enough, proven enough, and replicable enough that others want to be part of it. That attraction is also leverage. People join your effort, copy your model, align with your approach - and in doing so, they carry it further than you could alone. That is how Soft Power scales a solution that neither markets nor governments would fund.

Lijjat cannot force anyone to do anything. It has no political mandate and no financial leverage over other organisations.

And yet women keep joining. The cooperative has grown to 45,000 members across 82 branches in 17 Indian states not through recruitment drives or financial incentives, but because the model works and women can see it working. That is the attraction side.

The leverage side is replication. Multiple cooperatives across Indian states have modelled their governance on Lijjat’s system. Uganda sent a Vice-Presidential delegation to study it. Other international delegations have used its ownership structure as a development blueprint.

Nobody was asked, paid, or required. The model earned it. The Soft Power did the heavy lifting.

What Could Accelerate This

The world Lijjat operates in today looks nothing like 1959. That creates an opportunity it didn’t have before.

Lijjat’s quality control depends on centralised testing and physical infrastructure. That limits replication to places where Lijjat already has presence.

The opportunity is to turn that system into an AI-assisted protocol that any women’s cooperative anywhere could use - without Lijjat needing to be physically present. The governance model becomes something others can adopt and run independently.

The founding decision - self-reliance, collective ownership, and equal profit share - could reach a cooperative in Nairobi or Jakarta, running on a phone, without Lijjat ever being in the room.

To be clear, this is not about turning Lijjat into a technology organisation. It is about updating their leverage strategy to use available technology to extend their Soft Power further and faster - solving the Hard Problem beyond their current reach. And they can do this while remaining a profitable, self-reliant enterprise.

Why This Matters Now

Lijjat’s Hard Problem maps directly to SDG 5 - the UN’s goal to achieve gender equality and empower all women and girls by 2030. The current prognosis is not good. With only five years remaining until the 2030 deadline, none of the targets of SDG 5 have been fully achieved. If current trends continue, the world will reach 2030 with 351 million women and girls still living in extreme poverty. At the current pace, it will take 140 years for women to be equally represented in managerial roles, and 47 years to achieve equal representation in national parliaments.

The institutions designed to solve this are not moving fast enough. This is the gap where Soft Power can solve Hard Problems. And the conditions for it have never been better.

The tools that took Lijjat decades to build - governance architecture, replicable models, and trust at scale - can now be assembled faster. Digital infrastructure removes the need for physical presence. AI compresses timelines - the inspection, verification, and training that anchored Lijjat to specific geographies can now travel without it. Networks that once took generations to build can form in months.

The founding decision still matters. The commitment still has to be real. But the distance between a decision made on a rooftop and a model adopted in Nairobi, Jakarta, or Lagos is shorter than it has ever been.

That is what the Soft Power Index is built to document - and to make usable. To surface the pattern underneath these organisations clearly enough that someone else can build it faster to scale impact.

The questions are the same ones Lijjat answered in 1959. Today, there are new tools, technologies, and talent available to answer them.

Share

A Note on the Index

Lijjat is currently ranked 11th on the Soft Power Index - 7.9 on Soft Power, 8.3 on Hard Problem Focus. I didn’t find it. It was contributed by , who recognised it as the kind of organisation the index was built to surface.

This is the point of the Index being open. Organisations like Lijjat exist around the world - doing serious work, building proof at scale, yet rarely making it into mainstream leadership conversations, investment decisions, or founder strategies.

If you know one, tell us.

Leave a comment

Today is International Women’s Day. It is hoped that this article shows that women founders and leaders have always had the ambition and agency to build meaningful solutions to Hard Problems. What has changed is access to the tools to do it.

Lovable - the platform the Soft Power Index is built on, entirely by me - is free for 24 hours today. I am not a developer. The Index and everything you see at softpowerindex.lovable.app was built using Lovable (over the past week).

If you have an idea, today is a good day to start.

Hard Problems persist in the world - problems affecting hundreds of millions of people that markets won’t touch because they aren’t profitable, and that states can’t solve because they require commitments that cross borders, time horizons, and political cycles.

Some organisations are solving them anyway. They do it by building a different kind of power: not financial leverage, not political authority, but influence earned through commitment - the ability to change how other institutions behave through choices so specific, so costly to abandon, and so evidently effective that others voluntarily build on them.

The Soft Power Index exists to make that model visible, measurable, and imitable. It documents the pattern underneath the organisations that are succeeding, in enough detail that others can build on it.

What is Measured

Organisational Soft Power: this has two sides - and this index is concerned with both.

The first is Soft Power as outcome: the ability of an institution to change how other organisations and governments behave, through commitments and choices that others voluntarily adopt or build on. Not reputation. Not stated mission. Not the scale of operations or the size of the endowment. Demonstrated influence on external behaviour - specifically, whether others have changed what they do because of how this organisation has chosen to operate.

Evidence of this influence takes several forms: financial commitments that others have structured their own funding around; impact metrics verified by credible third parties; models replicated in new geographies without the founding organisation’s involvement; standards adopted into other institutions’ contracts; and the number and quality of organisations that have built on the model. These are not separately scored - they are the observable traces of the one thing being measured: whether this organisation’s choices have changed what others do.

The second is Soft Power as lever: the active use of that influence signal to accelerate impact - to convene coalitions, attract aligned capital, open institutional doors, and export a model without requiring the permissions, funding relationships, or political access that previous generations depended on. This is the dimension the index is designed to make actionable. The organisations profiled here did not just accumulate Soft Power - they deployed it. The founding decisions, the legible refusals, the costly commitments: these were not passive governance choices. They were active signals, sent deliberately, that changed what others were willing to do alongside them.

The index scores the outcome. The diagnostics, the Accelerant, and the leverage stack are designed to help founders and leaders activate the lever.

Why Replicability Is the Standard

Hard problems do not respect borders. They are structural, persistent, and shared across geographies - which means a response that stays local, or stays at the level of awareness, is by definition undersized relative to the problem it claims to address.

The index therefore asks not just whether an organisation has identified a real problem, but whether its approach has the reach architecture to match that problem’s scale. A replicable operating model is the mechanism by which Soft Power travels - it is how an organisation’s choices get adopted by others without coercion, without a marketing budget, and without a mandate.

This is the distinction between mobilising and governing. Mobilisation changes what people know or feel. Governance changes what institutions build. The index sits at the governance end. An organisation that names a problem with moral force, and builds a movement around it, may be doing essential work. But if there is no operating model that others can install - no protocol, no architecture, no structural arrangement that transfers - there is no Soft Power in the index’s terms.

The test is not “did others hear about this?” It is “did others build from this?”

Three Scored Dimensions

Soft Power (SP) - 45%

Does this organisation use influence, commitment, and moral authority - rather than financial leverage or political mandate - as its primary instrument for making progress on its Hard Problem? And has that model of operating been adopted by others who were not required to adopt it?

The highest scores go to organisations where Soft Power is both the method and the export: where the approach to the Hard Problem is itself built on earned influence, and where others have voluntarily built on that approach. Evidence of the method includes: outcomes achieved without coercive authority, coalitions formed through credibility rather than capital, and standards set by example rather than mandate. Evidence of export includes: models replicated in new geographies, standards embedded in other institutions’ contracts, and approaches that others have made their own without being required to.

Penalties apply for influence that is symbolic but not replicable; influence that derives from national or cultural prestige rather than the organisation’s own choices; and influence that is commercially convenient to claim.

Hard Problem Fit (HPF) - 35%

How specific, measurable, and genuinely difficult is the problem the organisation has named? The highest scores go to problems that are ignored by markets and states because they are unprofitable or politically inconvenient, defined precisely enough that progress can be verified by a credible third party, and not solvable by the organisation’s commercial self-interest alone.

The precision of the problem statement matters because vague problems permit vague commitments. Organisations that have named their Hard Problem specifically enough to be falsified are the ones making the kind of costly, specific choices this index is looking for.

Penalties apply for problems defined broadly enough to cover almost any activity, and for problems that the organisation benefits commercially from claiming to solve.

Capacity to Deliver (Cap) - 20%

Does the organisation have the track record, resources, and institutional infrastructure to deliver on its commitments? This dimension is weighted lowest because capacity without a Hard Problem and a real commitment produces nothing distinctive. A well-resourced organisation doing commercially convenient work is not what this index is documenting.

Index score

Index = (SP × 0.45) + (HPF × 0.35) + (Cap × 0.20)

Six Qualitative Filters

Before scoring, each candidate is assessed against six criteria that determine whether an organisation belongs on the index at all. These are the structural decisions that distinguish a commitment-led approach from a conventional one.

1. Soft Power as Primary Instrument
Does the organisation achieve progress on its Hard Problem primarily through influence, commitment, and moral authority - rather than through financial leverage, political mandate, or coercive authority?

This filter asks how the outcomes were achieved, not just what was achieved. An organisation with a large endowment that funds compliance is not exercising Soft Power - it is exercising financial power with a Soft Power aesthetic. An organisation that sets a standard others voluntarily adopt, builds a coalition through credibility rather than capital, or changes institutional behaviour through the force of a demonstrated commitment - that is Soft Power as instrument.

This is the filter most directly connected to leadership. The organisations on this index did not have mandates. They had clarity of purpose, costly commitments, and the discipline to hold both under pressure. That is what earned them the influence to move institutions they had no authority over. It is also what makes their model transferable - because influence built on moral authority travels in ways that financial leverage cannot.

Where this filter is not met, the organisation may still be doing important work. But it is not exercising soft power in the index’s terms - and the Soft Power score should reflect that accordingly.

2. Legible Refusal
Has the organisation declined something it could have profited from, yet which compromised its core commitment - and can the specific decision be named and dated? A commitment only has meaning if it has cost something. A refusal to charge for a drug that could be sold. A refusal to accept carbon credits that bypass national sovereignty. A refusal to take a management fee that would have compromised the cross-subsidy model. Where a legible refusal exists, it is the clearest early signal of a genuine commitment. Where it does not, the absence is not automatically disqualifying - not every organisation faces a profitable alternative to decline. The question then falls harder on the other filters, particularly costly commitment and model export.

3. Costly Commitment
Has the organisation made a structural commitment with observable consequences for violation - not a pledge, not a target, not a strategy document? A cross-subsidy model where free care is funded by paying patients, and therefore cannot survive if the commitment is abandoned, qualifies. An annual report promise reviewed by a board does not. The test is architecture, not intention.

4. Hard Problem Specificity
Can the problem be stated in one sentence with a credible third party able to verify progress? “Eliminate river blindness” is specific. “Improve global health” is not. “Develop a safe, effective vaccine within 100 days of identifying a new pandemic pathogen” is specific - and externally endorsed as a metric. If the problem description requires caveats, it is probably too broad to generate the kind of specific, costly commitment this index is looking for.

5. Model Export
Has the model been copied by others who were not required to copy it? This is the hardest test to pass and the most important when passed. A management methodology replicated in hundreds of hospitals across dozens of countries. An equitable access clause appearing in contracts at organisations that were never asked to include it. A care model for informal workers being adapted by governments that had no relationship with the founding organisation. Replication by choice is evidence of a working model - and the primary mechanism by which these approaches scale.

6. Honest Tension
Every actor on the index carries a tension - a place where its commitments come under pressure or its behaviour does not fully match its stated principles. If a tension cannot be named, the research is unfinished. The tension is included in the profile not to disqualify the actor, but because pretending that high-impact models are without flaw makes them less useful as models, not more.

The Practical Test

Every actor profile includes a Practical Test - three fields that translate the scores into concrete, specific decisions. Together they provide a past, present, and future reading of the organisation’s commitment.

What They Refused
The specific, costly thing the organisation declined - the profitable path not taken. This field identifies the legible refusal in practice: not as a principle, but as a named decision with a real consequence. BRAC refused to franchise the Graduation methodology without fidelity to design, declining partnerships that would have scaled faster but compromised the model’s integrity. Lijjat refused all donations and outside equity for over six decades, choosing to fail rather than be saved by charity. The refusal is the signal.

Who Copied Them
The evidence of model export: which governments, institutions, or organisations adopted this approach without being required to, and what specifically they built. This field is the primary verification of Soft Power. Forty-plus governments embedding the Graduation approach in national poverty programmes. The Sarvodaya cooperative model studied by international delegations and adapted by state institutions across India. If this field is thin, the SP score should be too.

Score Fall Trigger
The specific future event that would cause the score to drop - not a vague risk, but the precise structural change that would dissolve the commitment the score is built on. BRAC’s score falls if methodological discipline is sacrificed for scale. Lijjat’s score falls if the organisation accepts private equity or transitions to a hierarchical management structure. Anthropic’s score falls if it accepts a defence contract or allows national security carve-outs to erode its published commitments. The trigger is forward-looking accountability - it names what the organisation is actually protecting.

Win-Win, Not Extractive

A condition of inclusion on this index is that the organisation’s model produces outcomes where the benefit is shared with the populations it serves — not captured at their expense.

This is not a requirement for sainthood. Several profiled organisations operate in commercial markets or carry genuine contradictions. What is required is that the core commitment - the thing that earns the score - runs in the same direction as the interest of the people the Hard Problem affects. The eye hospital’s cross-subsidy model cannot survive without good outcomes for poor patients. The vaccine alliance’s equitable access clause is a condition of every contract, not a discretionary add-on. The commitment and the population benefit are structurally aligned.

Organisations whose influence is built on extracting value from the populations they nominally serve are not eligible for this index, regardless of their stated mission.

Scored Events

Each actor profile includes a log of specific, dated decisions that moved the score positively or negatively. Negative events - broken commitments, unresolved contradictions, failures of execution - are included, not omitted. All events are source-cited. Where a claim could not be verified through a primary or credible secondary source, it is not included.

Scores are updated when new events are verified. The index is alive, not a snapshot.

The Accelerant

The index documents what worked. But the timeline underneath most of these profiles — the decades between founding decision and demonstrated Soft Power - can be misleading. It suggests that this kind of influence is inherently slow to build. It isn’t. It was slow when these organisations built it, because the infrastructure that could have compressed the timeline didn’t exist yet.

Each actor profile therefore includes two additional fields that go beyond measurement.

The founding decision identifies the specific early structural choice that started the compounding clock - not the organisation’s full history, but the one decision that predetermined the trajectory. Merck’s “as much as needed, for as long as needed.” Aravind’s cross-subsidy architecture, framed as a business model rather than a charity. CEPI’s equitable access clause, written into the first contracts before any manufacturer had reason to accept it. These decisions were not obvious at the time. They are visible in retrospect precisely because they worked.

The acceleration lever identifies how digital infrastructure, AI, or network effects could compress the same journey for an organisation building this model now. A methodology that took fifty years to reach forty governments can travel as an openly licensed digital commons. A commitment that took decades to prove through consistent behaviour can be encoded as a smart contract, auditable from day one. A training programme that replicates one cohort at a time can become a product. The lever is different for each actor - because the founding decision was different, and the bottleneck is different. What is consistent is that the bottleneck exists, and that it is now removable.

The organisations on this index built Soft Power slowly because they had to. The argument of the Accelerant is that the next generation of organisations working on Hard Problems does not.

Using This Methodology

The six filters, three dimensions, and Practical Test above are designed to be applied, not just observed. If you are building or funding an organisation working on a Hard Problem, the most useful questions to ask are:

• Are we building toward influence or toward leverage? What is the primary instrument we are using to make progress - and if it is financial or political power rather than earned authority, what would it take to shift that?

• What specifically is the Hard Problem, stated in one sentence with a named verification metric?

• What is the legible refusal - the profitable thing we are choosing not to do?

• What is the costly commitment - the structural arrangement that would genuinely hurt to break?

• What would model export look like in this context - what would others copy, and what would they need to copy it?

• What is the honest tension - where does our model come under pressure, and what would we do if it did?

• What is the acceleration lever - the specific way that digital infrastructure, AI, or network effects could compress this model’s timeline, and what would need to be true to activate it?

The organisations on this index did not start with all of these elements in place. Most built toward them over years or decades. The pattern is visible in retrospect. The argument of this index is that it can also be built toward deliberately.

Leave a comment

The Soft Power Index is is part of Soft Power, Hard Problems, an advisory initiative by Sophie Krantz. Methodology questions and suggested additions can be directed to sophie [at] sophiekrantz.com.

In February 2026, Anthropic CEO Dario Amodei told the Pentagon something rarely said by an executive: we cannot in good conscience agree to your request.

The request was to relax AI safety guardrails - specifically the ones that prevent Claude from being used for mass domestic surveillance and fully autonomous weapons. The Pentagon wanted them gone. Anthropic said no. Trump’s administration responded by labelling Anthropic a supply chain risk and ordering federal agencies to phase them out.

A reported nine-figure contract was gone.

That decision has been on my mind. It’s somewhat unusual in 2026 for a company to have values. And, it’s very much unusual for a company to spend them. At a real cost. Under real pressure. In public.

So, I started building an index.

Soft Power is only real when you spend it

Soft Power: everyone claims to have it. Very few spend it on anything that matters.

Reputation, restraint, governance, and trust - these are some of the currencies of Soft Power. They accumulate slowly, through consistent behaviour over time. And like any currency, they’re only worth something when you’re willing to exchange them for something real.

Most institutions and leaders spend their Soft Power on themselves. On panels. On positioning. On maintaining access to the rooms they’re already in. The influence stays circular: it buys more influence, which buys more access, which buys more influence. Nothing moves.

The test - and really, the key test in 2026 - is whether an institution is willing to spend its Soft Power on a Hard Problem. A problem that affects millions or billions of people. A problem where the cost of action is real and the outcome is uncertain. A shared problem that no single actor can fix alone.

Anthropic recently passed that test. Not perfectly. Not conclusively. Yet visibly, at cost, and in a way that changes the terms for everyone else.

Subscribe now

What they did

The mechanics matter, because this is where Soft Power can be evaluated as strategic rather than sentimental.

Saying no was the visible part. The architecture underneath is what made it count:

• Their AI Constitution creates written red lines: no assistance with mass-casualty weapons, no enabling of mass surveillance, and no powering of fully autonomous lethal systems.

• Their Long-Term Benefit Trust gives a governance body the structural authority to defend those lines over decades, not just over a news cycle.

• Their internal Frontier Red Team sits inside the policy division and stress-tests the most advanced systems against exactly these risks.

These go beyond being traditional communications tools. Rather, they’re commitment devices - mechanisms that make the refusal believable. Walking the decision back would require dismantling the architecture - far beyond just changing a press release.

When the Pentagon called, Anthropic had already pre-committed. The Soft Power was structurally embedded.

That’s the difference between Soft Power as reputational nice-to-have and Soft Power as a strategic weapon.

Share

The Hard Problem they’re spending it on

The Hard Problem here isn’t AI safety as an abstract concept. The Hard Problem is two specific system-scale risks: the normalisation of mass domestic surveillance as a standard government capability, and the delegation of lethal force to machines without meaningful human control.

Both risks affect billions of people. Both are entangled with politics, markets, and technology in ways that no single actor can unravel. Both are moving fast enough that the norms being set right now - by labs, by governments, by procurement decisions - will be extremely difficult to reverse in ten years.

Anthropic’s refusal doesn’t solve either problem. More permissive competitors can fill the gap. The Trump administration is already rewarding them for doing so. The substitution risk is real, and anyone saying otherwise is selling a slimmed-down story than the evidence supports.

What the refusal does do: it changes what’s possible to say in every other room where these decisions are being made.

Regulators in Brussels, civil society actors in Seoul, procurement officers in Canberra - they can now point to a major vendor that refused, at scale, at cost. That’s a different kind of proof than a policy paper or a think-tank recommendation. It’s a data point in the global system.

Soft Power spent on a Hard Problem doesn’t fix the problem. It shifts the terms on which the problem gets fought.

Building an index

The Anthropic decision raises a question: who else is doing this? Which institutions - across tech, finance, philanthropy, multilaterals, alliances - are actually spending their Soft Power on Hard Problems, rather than on themselves?

I didn’t find a complete answer. So I started building one, with no-code on Lovable.

The Soft Power Index tracks actors across sectors on three dimensions:

1. Soft Power: governance quality, willingness to refuse, norm-setting behaviour, and structural credibility.

2. Hard Problem focus: clarity and seriousness about a specific global challenge: energy transition, critical minerals, AI regimes, supply chain resilience, or democratic institutions.

3. Capacity: the actual ability to act at scale: relationships, resources, convening power, and reach.

Scores are judgment calls, updated as events unfold. The rubric will evolve with critique. The aim is not to anoint heroes or shame villains. It’s to give leaders a diagnostic: if you want your organisation to matter on the world’s hardest problems, this is one way of asking whether your influence is actually pointed there.

Anthropic sits near the top of the current leaderboard. The Pentagon decision is one of the clearest recent examples of Soft Power spent deliberately, structurally, and at genuine cost on a system-scale risk. The Global South gap in their norm-setting is real. The substitution problem is unresolved. The scores reflect both.

The Soft Power Index is openly maintained. If you have evidence that changes the picture - a decision I’ve missed, a score that’s wrong, an actor that should be on the board - there’s a submission form on every profile. Open data, not open editing.

The question for you:

I built this for leaders whose decisions reach beyond their own organisations or current role. If you’re reading this, you probably already have Soft Power - reputation, relationships, governance, and trust - that you’ve accumulated over years of serious work.

The question the index is designed to force is: what are you spending it on?

Not what panels are you on. Not what access do you have. What Hard Problem - specific, system-scale, and affecting people you will never meet - is your influence pointed at?

Anthropic answered that question under maximum pressure. The answer cost them a nine-figure contract and their relationship with the most powerful government on earth.

That’s one answer. The index exists to find others.

Postscript - May 2026

The verified timeline of February 2026 is more complex than the original piece reflected. On 9 January 2026, Defense Secretary Pete Hegseth signed a memo directing that all military AI systems must allow “any lawful use.” On 14 February, the Pentagon publicly signalled it was considering ending its relationship with Anthropic. In the week of 24 February, Hegseth met Dario Amodei and delivered a formal ultimatum, with a deadline of 5:01 p.m. on 27 February. On 24 February, the same day the formal ultimatum and deadline were set, Anthropic published Responsible Scaling Policy v3.0, removing the binding commitment to pause development and deployment if it could not implement required safety mitigations before reaching the next capability threshold. The pause commitment had been the policy’s flagship feature since 2023. On 26 February, Anthropic formally refused the contract. On 27 February, the deadline passed and the ban fell.

The original piece, published 1 March, read the refusal as the proof of soft power as structural pre-commitment. The verified timeline complicates that reading. The visible boundary was defended at cost. The structural commitment underneath it was being loosened on the same day the ultimatum was delivered. Both moves were happening under the same six weeks of sustained pressure.

The framework still holds, but the case it scores is different. Soft power is only real when you spend it. The corollary, equally true, is that an institution can defend a visible decision publicly while quietly weakening the architecture that made the decision credible. The 2026 case is both at once. The Index scores reflect both: the cost of the public refusal, and the depreciation of the structural backbone that made it possible. SP -0.5, HPF -0.3.

The Index exists to track exactly this kind of complexity. The original argument stands.

Leave a comment

Shared.

Hard Problems: One sentence

Hard problems are shared global challenges - affecting millions or billions of people - where the risks and the rewards are connected across borders, and no single actor has the mandate, money, or technology to solve them alone.

Hard Problems: One paragraph

The most valuable problems left to solve are the ones everyone agrees are urgent and almost no one is structurally positioned to own. Hard problems are where this decade’s outsized opportunities live. In 2026, they show up in climate transition, critical mineral sovereignty, AI safety, biosecurity, demographic shifts, planetary health, and institutional trust. But not every large or complex challenge qualifies. A hard problem, in the precise sense, has a specific structure: it affects millions or billions of people, it can be stated in one sentence and verified by a credible third party, and it sits in the gap that markets will not enter because it is not sufficiently profitable and states cannot solve because it requires commitments that cross borders, decades, and political cycles. That gap - abandoned by both - is exactly where soft power actors have the most leverage. The organisations making the biggest moves here are not waiting for incentive structures to change. They are building the kind of influence that pulls others into the solution with them.

Why Soft Power and Hard Problems?
Soft power is the trust and influence an actor can deploy, and hard problems are the shared challenges where that effort can now unlock outsized strategic and commercial gains. They are the places where external leverage, coalitions, and new tools make solutions that once looked impossible suddenly actionable.

Why this pairing?
Because in 2026, the real test of soft power is which hard problem it is spent on and whether that choice turns influence into system‑level outcomes, not just reputation.

Read more on Soft Power in the link below.

---

Hard Problems: One page

Why hard problems now

The most valuable problems left to solve are the ones everyone agrees are urgent and almost no one is structurally positioned to own. Hard problems are where this decade’s outsized opportunities live. For leaders, boards, and founders in 2026, hard problems are no longer a CSR afterthought or someone else’s mandate; they are an emerging frontier of strategy, capital allocation, and survival. There are clear reasons for this, and there are practical ways to work with hard problems to generate outsized results.

Acting at system scale

Big moves are no longer available only to big players. Exponential technologies, AI agents, and global platforms mean even small, focused organisations or founders can now act at system scale. A well‑positioned board and a lean founding team can draw from essentially the same toolbox to tackle system‑scale problems. At the same time, climate risk, supply‑chain fragility, AI governance, and social instability now show up directly in revenue, cost of capital, and licence to operate. Treating these as externalities is an emerging strategic risk, rather than a neutral position.

Off‑track goals, on‑table opportunities

The UN Sustainable Development Goals (SDGs) are off-track, yet the opportunities to solve them are not. Global SDG assessments warn that most targets are unlikely to be met, with a significant share going backwards. In that context, inaction becomes an active decision to ignore massive unmet demand and systemic inefficiency. Those gaps sit inside systems designed for a different era, where risk and value are still mispriced. It is like coastal real estate being priced as if rising sea‑levels do not exist, or critical‑minerals supply chains priced as if geopolitical shocks do not happen. The mismatch between old assumptions and new realities creates some of this decade’s biggest commercial arbitrage opportunities.

Hard problems as a lens

Hard problems are already impacting business. Every major organisation is embedded in these systems, whether acknowledged or not: through where it sources, how it uses data and AI, which markets it depends on, and which communities it relies on for talent and legitimacy. Seen this way, hard problems are not insurmountable obstacles however they are a high‑resolution lens for strategy - a way to see where an organisation is most exposed, most essential, and most able to move the system. For some boards, that makes hard problems the natural home for one or two genuine moonshots rather than just another layer of incremental initiatives.

What makes a problem hard

Hard problems follow a recognisable pattern. A hard problem typically:

• spans multiple jurisdictions and sectors;

• has deep path dependence (today’s constraints were created by yesterday’s decisions);

• involves competing narratives of fairness and legitimacy;

• and is vulnerable to non‑linear shocks and feedback loops.

A hard problem also meets three precise conditions: it affects millions or billions of people; it can be stated in one sentence and verified by a credible third party; and it sits in the gap that markets will not enter because it is not sufficiently profitable, and states cannot solve because it requires commitments that cross borders, decades, and political cycles. Problems that fail any of these tests may be important and urgent - but the leverage available here is specific to problems with this structure.

Single fixes do not move these. They cannot be resolved with a single product, a one‑off capital project, or a press release. They require a portfolio of interventions over time: standards and incentives, technology and infrastructure, coalition‑building, and narrative shifts. Hard problems live at the intersection of hard power (capital, mandates, and assets) and soft power (trust, legitimacy, and convening power). Leaders who see only one side either over‑invest in blunt force or over‑index on talk; the work is in combining both.

Who hard problems are for

Hard problems are not for everyone. They tend not to suit leaders whose primary motivation is maximising short‑term financial returns or building a tech‑led business for its own sake. The leaders who increasingly gravitate here - in companies, institutions, or as founders - usually care deeply about the underlying problem, not just the product, valuation, or status that comes with working in the space.

They often bring soft power as well as hard skills: the ability to work across cultures and countries, to convene people who would not otherwise sit at the same table, and to advance the work without needing to be the star. Many are also impatient with the status quo: tired of watching hard problems go unsolved because of inaction, constrained budgets, or the failure to deploy the best tools and technologies that already exist. Those are the leaders for whom hard problems become a natural home, rather than a side interest or a rung on a career ladder.

Setting an advantage in the year ahead

Four dynamics are shaping how hard problems set an advantage in 2026:

1. From cost centre to strategic optionality

Hard problems are changing where they sit on the balance sheet. Historically, boards put them into buckets such as regulatory risk, sustainability, or reputation management - all important, yet conceptually separate from growth and competitive advantage.

The emerging move in 2026 is to treat them as sources of strategic optionality: new markets, new products, and new forms of resilience that appear when an organisation is designed to solve part of a system‑scale problem. Decarbonising a supply chain may begin as compliance, and it can also be a play for advantaged access to future customers, talent, and green capital. Building trustworthy, well‑governed AI may be booked as cost, and it can also become the entry ticket to high‑trust markets where opaque players are being legislated out.

In this frame, hard problems become the testing ground for an organisation’s ability to generate optionality under uncertainty, not a drag on earnings. This is where the few genuine moonshot‑scale bets live as deliberate 10× plays, not as side projects. The same logic applies from early‑stage ventures to listed companies: AI‑first startups are increasingly built around solving a slice of a global challenge from day one, using exponential tools that were previously reserved for incumbents.

2. Orchestrator vs builder

The focus is shifting from building things to orchestrating them. Working on hard problems no longer requires becoming a climate‑tech lab, AI research centre, or health‑tech giant; many powerful technologies are now accessible on demand and the barriers to accessibility continue to decrease.

Walmart’s Project Gigaton is an example. Rather than building all the solutions internally, Walmart set a goal to eliminate 1 billion tonnes of supply‑chain emissions by 2030 and then orchestrated thousands of suppliers, NGOs, and data tools to change how products are made and moved, using standards, calculators, and incentives instead of owning every asset or technology.

A founder‑led example looks similar in spirit, just on a different scale. Think of an early‑stage climate or health startup that assembles an ecosystem instead of trying to do everything in‑house: using off‑the‑shelf AI models and cloud infrastructure, partnering with universities for research, plugging into existing logistics networks, and coordinating corporates, regulators, and community groups around a specific hard problem mission. The value comes less from owning each component and more from seeing the system clearly, setting the direction, and getting diverse actors to move together.

Many leadership conversations, asked in traditional industries, ask “Do we want to be a tech company?”. An alternative question, which can yield greater insights, is: “Which system are we trying to shift, and what is our role in orchestrating the partners, tools, and capital to do it?”. In that role, boards and founders set direction, shape standards, and stitch ecosystems together rather than owning every component. When infrastructure and hard assets are increasingly available as services, the real constraint is less about technology and more about strategic focus and execution.

This is the governance shift behind mission‑oriented programmes and serious moonshots: treating hard problems less as engineering tasks and more as orchestration challenges. For founders, the primary moat is often not proprietary technology but soft power - the ability to convene talent, capital, and partners around a credible hard problem mission, and to build influence that pulls others in rather than leverage that forces them.

3. Hard problems as the ultimate stress test

Hard problems are a live diagnostic. Used deliberately, they reveal how leadership, culture, and governance actually perform under constraint.

To see this in action, set a leadership team a concrete challenge - for example: stand up a critical‑minerals recycling loop across three continents, take 40% of emissions out of your freight network, or deploy safe AI copilots across every major customer journey - and constrain it to existing assets and tight budgets.

What happens next is revealing:

• Who reaches first for hard spend - more headcount, new capital, or a re‑org.

• Who mobilises soft power - existing coalitions, standards, narrative, or partnerships.

• Who recognises the organisation’s soft power balance sheet - with a plan to leverage trust, relationships, and convening power as a usable asset.

Handled this way, hard problems become a stress test of strategic imagination, internal alignment, and orchestration capability. They expose whether an organisation can sustain ambition, learning, and coalition‑building over years.

A useful question follows: who else has this hard problem, and where in the world is it already being solved or seriously tackled? Instead of assuming the full solution must be built in‑house, boards and founders can ask whether lower‑cost, lower‑risk, more scalable models already exist that could solve it for multiple players, not just one organisation. That additional dimension - estimating the time and cost to solve the problem system‑wide, not just within a single footprint - creates a clearer choice: is this a mission to own, or is there more leverage as a reference client and scale partner for those solving it globally?

4. Why the board or founder should care

Hard problems are an organising principle as well as a risk category. They offer an efficient lens to test whether strategy, risk appetite, and culture are suited to the next decade.

They force zoom‑out questions:

• Which systems truly underpin the business?

• Where is the organisation a price‑taker - and where could it be a standard‑setter?

• Which coalitions is it part of by accident rather than design?

• Could a billion‑dollar business be built by solving a global challenge in this system?

• Could a billion people be positively affected if this mission actually delivered?

They also guide zoom‑in decisions:

• Which experiments to fund.

• Which technologies to adopt.

• Which markets to exit.

• Which narratives to own.

• Which under‑used soft power assets - trust, relationships, and convening power - could be mobilised to support outsized results.

A board that can articulate its hard problem thesis is often ahead of competitors reacting piecemeal to regulations, headlines, or activist pressure. Maersk is an example: its thesis is that decarbonising global shipping is a non‑negotiable hard problem, so it has committed to net‑zero emissions by 2040 and is investing in green‑methanol vessels and long‑term fuel offtake agreements, while many rivals still treat climate rules as intermittent compliance events.

Hard problems, framed this way, become an organising principle for serious leadership. They align portfolio, technology bets, partnerships, and culture around a small number of system‑relevant missions - including one or two clearly defined moonshots where a 10× outcome would materially shift strategic position, even if the path includes productive failure and spillovers. The same lens that helps a board distinguish price-taker from standard-setter helps a founder decide which arena to play in - and how to use a high-trust footprint to build the kind of gravitational pull that makes others’ success dependent on yours.

Soft Power, Hard Problems

Soft power is the fuel; hard problems are where it’s now worth spending it.

This piece is part of an ongoing exploration of how hard problems and soft power can give founders and leaders a different way to play the next decade. What are you seeing, experimenting with, and achieving when you use soft power to solve hard problems? What wins are currently out of reach that you’d like to materially edge closer to this year (without emulating zero-sum tactics)?

Leave a comment

How leaders can engage China to solve hard problems faster, easier, and with greater scale, and the soft power plays that make this possible:

Further Reading

Hard problems, missions, and moonshots

• Mazzucato, M. (2018). Mission‑oriented research & innovation in the European Union: A problem‑solving approach to fuel innovation‑led growth.
  https://op.europa.eu/en/publication-detail/-/publication/5b2811d1-16be-11e8-9253-01aa75ed71a1/language-en

• European Commission – Mission‑oriented R&I policies: in‑depth case studies (Apollo, others).
  https://research-and-innovation.ec.europa.eu/knowledge-publications-tools-and-data/publications/all-publications/mission-oriented-research-and-innovation-policy-depth-case-studies_en

• Krantz, S. (2023). “What is the global problem you could solve from where you are?”
  https://open.substack.com/pub/sophiekrantz/p/what-is-the-global-problem-you-could?r=wpxa&utm_campaign=post&utm_medium=web

• PNAS (2023). “The macroeconomic spillovers from space activity.”
  https://www.pnas.org/doi/10.1073/pnas.2221342120

• NASA History. “Managing America to the Moon: A Coalition Analysis.”
  https://www.nasa.gov/history/SP-4219/Chapter8.html

SDGs and why hard problems are worsening

• UN Statistics Division (2025). The Sustainable Development Goals Report 2025.
  https://unstats.un.org/sdgs/report/2025

• UN Statistics Division (2023). The Sustainable Development Goals Report 2023: Special Edition – Progress at the Midpoint.
  https://unstats.un.org/sdgs/report/2023

• World Economic Forum (2025). “Sustainable Development Goals: Are we on track for 2030?”
  https://www.weforum.org/stories/2025/09/sdg-progress-report-2025

• Earth.org (2024). “Only 16% of SDG Targets on Track to Be Met by 2030, Report Finds.”
  https://earth.org/only-16-of-sdg-targets-on-track-to-be-met-by-2030-report-finds

• Amnesty International (2025). “Why are the Sustainable Development Goals way off track?”
  https://www.amnesty.org/en/latest/campaigns/2025/06/why-are-the-sustainable-development-goals-way-off-track

Exponential tech, agentic AI, and small‑unit agency

• World Economic Forum (2025). From Safer Cities to Healthier Lives: The Top 10 Emerging Technologies of 2025.
  https://www.weforum.org/stories/2025/06/top-10-emerging-technologies-of-2025

• World Economic Forum (2026). Technology Convergence Report 2025 (digest).
  https://www.weforum.org/publications/technology-convergence-report-2025/digest/

• Google Cloud (2025). “5 ways AI agents will transform the way we work in 2026.”
  https://blog.google/innovation-and-ai/infrastructure-and-cloud/google-cloud/ai-business-trends-report-2026

• DeepL / Press (2025). “69% Global Executives Predict AI Agents will Reshape Business in 2026.”
  https://www.deepl.com/en/press-release/69_global_executives_predict_ai_agents

• Microsoft WorkLab (2026). “Agents are here - is your company prepared?”
  https://www.microsoft.com/en-us/worklab/agents-are-here-is-your-company-prepared

• Rezolve.ai (2026). “One Billion AI Agents by 2026: What This Means for ITSM and Beyond.”
  https://www.rezolve.ai/blog/one-billion-ai-agents-by-2026

Boards, founders, and soft power

• Brand Finance (2026). Global Soft Power Index 2026.
  https://static.brandirectory.com/reports/brand-finance-soft-power-index-2026-digital.pdf

• World Economic Forum (2026). “Next‑generation leadership can help move the world forward.”
  https://www.weforum.org/stories/2026/01/leaders-wef-trust-ygl-new-kinds-of-leaders

• FasterCapital (2025). “The Soft Power in Startup Leadership Development.”
  https://www.fastercapital.com/content/The-Soft-Power-in-Startup-Leadership-Development.html

• Startup Genome (2025). Global Startup Ecosystem Report 2025.
  https://startupgenome.com/contents/report/gser-2025_4786.pdf

• Innovations Venture Studio (2025). “The Rise of AI‑First Startups: What 2025 Will Look Like.”
  https://innovationsventure.studio/the-rise-of-ai-first-startups-what-2025-will-look-like