There is something unusual about the economic future of the next fifteen years.

Many of its most important demographic features are already known.

We do not know exactly how artificial intelligence will transform employment, whether interest rates will remain elevated or how governments will respond to growing fiscal pressure.

But we know that the large generations born after the Second World War are retiring.

We know that working-age populations are declining across much of the developed world.

And we know that even a substantial increase in birth rates today would not significantly change the number of people available for work during the 2030s.

This creates a distinction between the demographic future, which is relatively predictable, and the economic response, which remains uncertain.

That distinction is the foundation of the final article in our Shrinking Core research series.

Over the previous three articles, we examined why population totals no longer provide an adequate picture of economic capacity, how ageing economies can adapt their production models, and why the same AI technologies can produce very different outcomes across countries.

Now we bring those forces together.

The question is no longer simply what ageing will do to economies.

It is what individuals, businesses and capital owners should do while there is still time to adapt.

Germany's working-age population is projected to decline by between 3.2 and 6.2 million by 2035, depending on migration.

Japan's population aged 65 and over is expected to approach 35% by 2040.

South Korea's working-age population is projected to fall from 36.6 million in 2023 to 27.2 million by 2044.

China's population aged 16–59 declined by approximately 6.6 million in 2025 alone.

These figures describe different countries, but they point toward the same structural transition.

A smaller proportion of society will be responsible for producing goods and services, generating taxable income and supporting an increasing number of older residents.

The adjustment will not necessarily produce a sudden crisis.

More often, it will appear through incremental changes that are easy to underestimate when considered separately.

Pension contributions increase.

Retirement ages rise.

Employers struggle to recruit skilled workers.

Local governments consolidate services.

Some housing markets lose demand while others remain expensive.

The important analytical point is that these developments reinforce one another.

Throughout this series, we have used the concept of the Economic Core to distinguish population size from actual economic capacity.

The Economic Core consists of residents who simultaneously participate in productive employment, earn sufficient income, contribute positively to public finances and support domestic consumption.

Its capacity can be understood through three connected factors:

Economic Core capacity = Core size × Core productivity × Core capture

Core size represents the number of productive residents.

Core productivity represents the value each person creates, supported by skills, technology and capital.

Core capture represents how much of that value reaches households and remains within the domestic economy.

Demographic change places growing pressure on the first variable.

Artificial intelligence, automation and capital investment may substantially improve the second.

But the third will determine whether technological progress translates into stronger wages, household demand and fiscal sustainability.

This is why an economy can become more productive without becoming more resilient.

Higher output does not automatically mean higher household incomes.

And higher GDP does not necessarily indicate that the productive foundation supporting public finances is becoming stronger.

One of the clearest examples comes from Germany.

Its statutory pension contribution rate has remained at 18.6% since 2018.

The country's pension insurer projects that the rate may need to rise to 19.9% in 2028 and approximately 20% in 2029 as reserves decline.

The mechanism is straightforward.

Pay-as-you-go pension systems depend on contributions from current workers to finance benefits for current retirees.

As the ratio between those groups changes, governments must adjust contributions, benefits, retirement ages or financing from general taxation.

Higher contributions help sustain the system, but they also increase employment costs and reduce disposable income.

That creates a difficult trade-off.

Ageing economies need skilled workers to remain economically productive, yet financing retirement obligations can make employing those workers more expensive.

Japan illustrates another dimension of the problem.

Its ten-year government bond yield reached approximately 3% in October 2026.

For many years, Japan appeared to demonstrate that ageing societies naturally experience low inflation and low interest rates.

But demographic change can also produce labour shortages, higher wage pressure and greater fiscal obligations.

Whether ageing ultimately produces higher or lower real interest rates remains contested.

The practical implication is that households, companies and investors should not build long-term plans around only one interest-rate regime.

Demographic change and financing costs interact particularly strongly in property markets.

A country can experience population decline while its strongest metropolitan areas continue attracting workers, students and capital.

At the same time, peripheral regions may lose residents, schools, businesses and essential services.

Japan already has approximately nine million vacant homes.

Tokyo's population is projected to peak around 2030, with its central wards beginning to decline after 2035.

These developments do not mean that all Japanese property will lose value.

They mean that the difference between locations may become increasingly important.

A home in an economically resilient district with strong employment, transport and healthcare access has a different long-term demand profile from one in a region experiencing persistent population decline.

For households, this changes how property should be evaluated as a retirement asset.

For businesses, it affects where facilities, stores and service networks should be located.

For capital, it means that real estate should be analysed at the level of specific markets rather than treated as a uniform asset class.

Governments are already trying to increase labour-market participation.

The European Union's employment rate among people aged 20–64 reached 76.1% in 2025, its highest level since the series began in 2009.

Japan has 9.43 million workers aged 65 and over, representing 13.8% of total employment.

Germany is gradually increasing its statutory retirement age to 67.

These changes matter.

They increase the number of people contributing to economic activity and public finances.

But their capacity to offset demographic decline is limited.

Consider Germany.

Its working-age population could decline by 3.2–6.2 million by 2035.

Even an ambitious increase in labour-market participation equivalent to three percentage points would add roughly 1.4 million workers.

That would compensate for less than half of the decline under the more favourable demographic scenario.

Participation can slow the contraction.

It cannot fully reverse it.

The remaining adjustment must come primarily from higher productivity and effective migration.

And even migration is not simply a matter of increasing arrivals.

The economically relevant question is how quickly newcomers become productive, self-sufficient participants in the labour market.

Artificial intelligence and automation offer a potentially powerful response to labour scarcity.

If a smaller workforce can produce more per person, an ageing economy may maintain output and living standards despite demographic decline.

But this outcome depends on two conditions.

First, businesses must adapt their production models.

That means investing in technologies that complement skilled workers, improving organisational processes and preserving essential technical capabilities.

Second, productivity gains must reach households.

If higher output translates into rising real wages, workers can sustain consumption and public finances.

If gains are concentrated among capital owners, domestic demand may weaken relative to productive capacity.

This distinction separates two very different economic futures.

One is a society with fewer workers who are more productive, better paid and capable of sustaining a high standard of living.

The other is a society with increasingly productive companies but stagnant household incomes and rising fiscal pressure.

Both are possible.

The difference will be determined by institutions, business models, labour markets and policy choices.

Our research identifies three broad scenarios for the rich ageing economies and China.

Base scenario — 55%: Two-speed adaptation

Economies adjust unevenly.

AI and automation increase output per worker, while real wages improve modestly in sectors facing labour shortages.

Pension contributions and retirement ages rise gradually.

Regional differences become more pronounced, with strong metropolitan areas retaining demand while some peripheral regions contract.

Most economies manage a gradual adjustment rather than a sudden breakdown.

Stress scenario — 30%: Fiscal squeeze

Productivity gains increasingly favour capital owners.

Household incomes fail to keep pace with output.

Pension and healthcare costs continue rising, while higher borrowing costs place additional pressure on public finances.

Governments increase taxes and contributions, weakening disposable income.

Property markets in declining regions become less liquid.

Extreme scenario — 15%: Disorderly break

A major ageing economy experiences a serious bond-market or pension-funding crisis in which demographic costs are central.

Alternatively, a sharp correction in AI-related investment coincides with trade fragmentation and a broader global slowdown.

The resulting adjustment is more abrupt, with significant consequences for asset prices, investment and employment.

These probabilities are analytical judgments rather than statistically calibrated model outputs.

Their purpose is to structure uncertainty and identify the evidence that would support or weaken each scenario.

A useful forecast should have a defined outcome, a measurable threshold and a date when it can be evaluated.

Our final article records four such forecasts.

Forecast 1 — Japan, September 2027

We assign a 75% probability that Japan's ten-year government bond yield will be at or above 2.5% on 30 September 2027.

The forecast will be evaluated using Japanese Ministry of Finance bond-yield data.

Forecast 2 — Germany, 2029

We assign a 75% probability that Germany's statutory pension contribution rate for 2029 will be set at 19.9% or higher.

The forecast will be evaluated using official pension-insurance and federal regulatory data.

Forecast 3 — European Union, 2030

We assign a 35% probability that the EU employment rate among people aged 20–64 will reach at least 78% for 2030.

The forecast will be evaluated using Eurostat data published by June 2031.

Forecast 4 — Japan, October 2035

We assign a 55% probability that Japan's total population on 1 October 2035 will be below 117 million.

The forecast will be evaluated using official population estimates published by June 2036.

These forecasts are designed to be falsifiable.

If the data contradict them, the analytical framework must be updated.

Rather than attempting to predict every economic development, our research identifies seven indicators that reveal how the structural transition is unfolding.

The first is the change in working-age populations across Germany, Japan, South Korea and China.

The second is whether real wages rise alongside productivity.

The third is labour's share of economic output, which helps reveal whether technological gains are reaching workers.

The fourth is pension contribution rates and retirement-age decisions.

The fifth is the relationship between government bond yields and nominal GDP growth.

The sixth is labour-market participation and the speed at which migrants enter productive employment.

The seventh is the widening or narrowing gap between economically resilient metropolitan areas and declining regions.

No single indicator provides a complete answer.

Together, they offer a more useful picture of economic adaptation than GDP growth alone.

The first priority is to understand where your skills fit within a changing labour market.

Over the next six months, examine whether demand for your profession is likely to increase as populations age.

Consider whether AI can make you more productive or whether it is likely to automate the routine tasks on which your work depends.

Over the following year, develop practical competence with the AI tools used in your field, including the ability to verify their outputs.

Technical maintenance, engineering, healthcare, infrastructure services and specialised advice for older clients are among the areas that may benefit from structural labour scarcity.

Housing decisions require a similarly long-term perspective.

Before buying property, examine the demographic and economic prospects of the specific district over fifteen to twenty years.

Look at the number of younger residents, local employment opportunities, housing vacancy rates and municipal finances.

Retirement planning should also account for potentially later retirement, higher contributions and changing interest-rate conditions.

The objective is not to predict one exact future.

It is to avoid decisions that depend on the continuation of conditions that are already changing.

Businesses should begin by reviewing the demographic composition of their customer base.

Which products depend on young families and new household formation?

Which serve older households, infrastructure maintenance or essential services?

Within the next twelve months, companies should evaluate whether their product mix matches the population structure expected in 2030 and 2035.

Workforce planning also needs to change.

In labour-scarce markets, automation should focus on increasing output per employee rather than simply reducing headcount.

Experienced workers may become more valuable as the supply of replacements declines.

Businesses should also evaluate the cost of financing under different interest-rate scenarios.

Supply-chain resilience deserves attention, particularly where strategic inputs depend heavily on a single country.

Finally, owners of small and medium-sized companies should consider succession planning.

Ageing economies face a growing risk of viable businesses closing because their owners retire without successors.

Documented processes and well-designed automation can make those businesses easier to transfer or sell.

The coming transition creates long-term opportunities, but structural demand must be distinguished from attractive valuations.

Industrial automation, infrastructure renewal, healthcare, senior housing and home adaptation are areas where demographic change may support sustained demand.

However, the performance of individual investments will still depend on competition, financing costs, execution and purchase price.

Property requires particular care.

Strong metropolitan areas with durable employment and service networks may behave very differently from peripheral regions with declining populations.

Investors should also consider how their portfolios would perform under both higher and lower interest-rate regimes.

Concentration in AI-related assets deserves scrutiny, particularly if market valuations assume investment returns that have not yet been demonstrated.

The key is to identify where technology and demographic change create durable economic value rather than relying on broad narratives.

Several developments would require us to revise the analysis.

If productivity in Japan and Germany grew above 2% annually for three consecutive years while real median wages kept pace, the probability of successful adaptation would increase.

If Japan's ten-year bond yield returned below 1% and remained there with inflation near target, concerns about a structurally higher interest-rate regime would weaken.

If the EU reached its 78% employment target by 2030 while maintaining productivity, labour-market participation would be proving more effective than our central estimate suggests.

A sustained recovery in fertility across East Asia would improve the outlook beyond the 2030s, although it would have little immediate effect on the workforce.

And if migrants entered productive employment substantially faster, migration could offset more of the demographic contraction than current trends imply.

The purpose of identifying these conditions is to ensure that the research remains responsive to evidence.

The demographic trajectory toward 2040 is largely established.

Many advanced economies will have fewer working-age residents, more retirees and growing obligations to finance pensions, healthcare and public services.

But demographic decline does not automatically mean economic decline.

A smaller workforce can support a prosperous society if it becomes more productive, receives an appropriate share of the value it creates and operates within institutions capable of adapting.

The central risk is not simply that there will be fewer workers.

It is that economies will continue making decisions based on the assumption that workers, young consumers and inexpensive financing will remain abundant.

The next fifteen years will reward different capabilities.

Skilled labour will become more valuable in many ageing economies.

Reliable infrastructure will become more important as the cost of maintaining dispersed networks rises.

Businesses will need to serve changing household structures.

And capital will need to distinguish productive adaptation from attempts to preserve economic arrangements that no longer fit demographic reality.

For individuals, the practical task is to develop scarce skills and make resilient decisions about housing, careers and retirement.

For businesses, it is to redesign production and customer strategies around the population that will actually exist.

For capital, it is to identify durable demand while remaining disciplined about financing costs and valuations.

The future is not fully determined. But many of the pressures shaping it are already measurable, and the cost of ignoring them increases with every long-term decision.

THRIVE IN CHAOS

Series I — The Shrinking Core

Article 4 of 4: 2030, 2035, 2040 — Forecasts and a Playbook for the Shrinking Economic Core

Full research: thriveinchaos.ai

Signal → Meaning → Action → Stability

Strategic intelligence for decisions under uncertainty.

This publication provides strategic analysis and is not individual investment, legal or financial advice.