The governance failure
The gradual submersion of Jakarta is not a climate story – it is a governance failure of the adaptation response.
The science was clear by the 1990s. 40% of the city sits below sea level. The worst districts sink twenty-five centimetres a year, not because of the sea or climate, but because of groundwater extraction: a political failure dressed as a natural disaster. The decision to leave came in 2019. Construction of Nusantara, Indonesia’s newly planned capital city, began in 2022. Then in 2025, President Prabowo reclassified it from national capital to political capital. State funding fell from $2 billion in 2024 to $300 million in 2026. The city built to house 1.2 million people by 2029 currently holds ten thousand.
Jakarta is sinking because of misalignment: political tenure is four years, while sea-level risk is measured in decades. Every actor behaved rationally within their window. Yet the aggregate of rational decisions is a rapidly sinking city and a trapped population.
The standard response to any critique of adaptation finance is: give us better data, longer horizons, and more sophisticated instruments. However, climate adaptation is predicated on a series of unknowns, including tipping points that we cannot predict. In short, we will not be able to mathematically model ourselves out of climatic shocks. What we need is a completely different class of financing and policy decision architecture, one that will allow us to harness the scale of one policy problem to fund the solution to another.
With the right reframing, African governments, financiers, and developers can collectively reorient themselves to approach climate adaptation and the associated possibility of migration as a large-scale opportunity for sustainable infrastructure development, powered by domestic construction industries.
The known unknowns
In April 2026, as NPR reported Nusantara as a city of doubt, J.P. Morgan published Tipping Points: Decision Making Under Deep Uncertainty. The coincidence is instructive. J.P. Morgan’s central finding was that climate tipping points sit in Knightian uncertainty. In other words, this is not calculable risk, but a domain where standard tools break down. Discounted cash flow (DCF) models with three-to-five-year forecast periods structurally misprice what is coming. We cannot accurately calculate or realistically establish outcomes to evaluate decisions.
The report recommends scenarios and tabletop exercises: the right tools for deep uncertainty. What it does not do is ask whether a different decision architecture might make the timing of the tipping point irrelevant.
The implication is clear. If the tool that prices risk cannot price this risk, the tool is simply wrong. The analytical path to climate adaptation is closed.
However, we have been here before – in deep uncertainty – for mitigation. China provided a response to that challenge.
China did not wait for certainty before acting on decarbonisation. It identified a simultaneity: the scale of renewable deployment would be the engine of domestic industrial policy. It did not solve decarbonisation and industrialisation sequentially. It used the scale of one to fund the other. Solar panels, electric vehicles, battery storage – each sector built on guaranteed domestic offtake before it competed globally. The result was the fastest manufacturing cost-curve descent in history.
Ultimately, China did not need to know when peak oil demand would arrive. It only needed to know that it would, and that the scale of its response would determine who manufactured its own transition.
Adaptation economics has not learned this lesson. Many have called it incomplete; J.P. Morgan suggests it is unknowable. Regardless of both epistemic criticisms (incompleteness and deep uncertainty), adaptation is still pricing and financing each seawall against each flood scenario, instrument by instrument – piecemeal. This is mitigation’s original sin, now applied to adaptation and resilience. The answer is to change the scale of the response until the valuation problem becomes moot.
Lifebelts and lifeboats
What we have today is in many respects a “lifebelt” climate adaptation architecture. Seawalls, resilience bonds, parametric insurance are all built on one assumption: the population stays, the assets are worth defending. Adaptation finance prices the cost of holding a location. It has no instrument for leaving one – the lifeboat. This is the incompleteness problem. J.P. Morgan says that tipping points mean that we cannot determine what is required – whether a lifebelt or lifeboat – until certainty arrives. By then, it will be too late.
Africa’s seven largest coastal cities will grow 40% by 2030, adding twenty-one million people to already-exposed coastlines. Permanent flooding of parts of Lagos, Cotonou, Dar es Salaam, and Alexandria is projected under mid-range scenarios by 2050. Meanwhile, the blue economy these cities anchor is on course to grow from $296 billion to $405 billion by 2030. The adaptation finance directed at protecting it is not operating at the same scale. Simultaneously, the receiving city – wherever displaced populations relocate – remains entirely unfinanced.
If we can transition away from simply holding the line towards executing a managed retreat, we may be able to produce a fiscal and developmental catalyst.
The unknown knowns
Every coastal city that crosses the threshold from defence to retreat can generate a construction demand event. New settlements require modular construction, climate-adapted urban design, water management, distributed energy, and digital infrastructure. African governments that structure that demand domestically, using managed retreat as the anchor offtaker, would be applying the same structural logic China applied to decarbonisation.
When done incorrectly, the receiving city is built with imported technology and imported supply chains. In this case, the fiscal multiplier exits, and the currency crisis enters. Africa gets the displacement without the development.
The Jakarta case is instructive precisely because Nusantara is being built on foreign capital. In fact, what the city is built from, and by whom, was never central to the decision. Nusantara is a national migration project dependent on the confidence of strangers in Indonesia’s ability to repay their investments in full and on time. There is no simultaneity; there is no synergy. There is only the familiar and fragile sequencing under pressure, as budgets, ambitions and timelines are scaled back.
Who builds, who pays
China scaled through one state, market and industrial policy; Africa must scale outward from cities and national markets. The test for Africa is whether this urban demand builds its domestic industry, construction capacity and local-currency finance. Unfortunately, most projects fail.
Senegal’s Diamniadio and Kenya’s Konza are locally owned or initiated, but both rely heavily on external finance, foreign contractors and engineering. In Nigeria, Enyimba Economic City comes closest in industrial intent, yet its first phase remains dollar-financed. Eko Atlantic is the clearest partial exception: its developer’s group produces concrete, aggregates and glass, although reclamation was foreign-contracted. Across these cases we see a consistent pattern: domestic actors own the city; foreign actors finance and build its highest-value components. Closing the gap requires reversing that hierarchy: domestic materials, contractors and currency at the core; foreign expertise only where and when necessary.
In Africa, current adaptation strategies are predominantly centred on lifebelts and not lifeboats. The window of opportunity is not defined by when the tipping point arrives, but rather by whether the question of building is answered before or after our cities – like Jakarta – sink under the sea. Will we defend or will we strategically retreat?
In this case, the real loss is not in abandoning ship. Rather, it is in building the wrong ship entirely.