When the population changes, the assumptions behind growth, interest rates and investment returns have to change too.

Most economic models quietly assume that the population they describe will keep behaving like the one they were built on. For much of the post-war period, that was a fair assumption. Working-age populations were growing across many rich countries. More people joined the labour market, incomes rose, and governments could count on a large workforce to help pay for public spending.

That is changing. Across Europe, Japan and China, people are living longer, birth rates have fallen and the share of older people is rising. In the EU, 22% of the population was aged 65 or over at the start of 2025, and the median age was 44.9 years (Eurostat). Japan’s share reached 29.4% in September 2025 (Statistics Bureau of Japan). China’s population fell by 3.39 million in 2025 to 1.40489 billion; 15.9% were aged 65 or over (National Bureau of Statistics). In some economies the working-age population is already shrinking. Others will follow in the coming decades.

That matters for economic growth, but it reaches further. Ageing population changes how households save and spend, how firms invest capital, how governments pay for pensions and healthcare systems, and possibly how financial markets price capital. So here is the question we keep coming back to: what happens when the demographic assumptions inside our models stop being true?

The figures already show the scale of the shift. Japan is close to three older residents in every ten; the EU is above one in five. China’s smaller percentage still represents more than 223 million people. These are different demographic transitions, not interchangeable economic outcomes. Labour-force participation, migration and productivity will decide how much pressure each economy actually faces.

The Economic growth problem starts with labour

Economists often use a simple production function, the Cobb-Douglas form:

Yₜ = Aₜ Cₜᵅ Lₜ⁽¹⁻ᵅ⁾

Here Y is output, A is total factor productivity, C is the capital stock, L is labour input and α is capital’s output elasticity (let us assume 35% in the example). Taking logarithms and differentiating with respect to time gives the continuous-time growth equation:

gY = gA + αgC + (1 − α)gL

Rearranged, it tells you how much productivity growth an economy needs to hit a target:

gA = gY − αgC − (1 − α)gL

Here is what that means in numbers. Say for example, an economy wants 2% output growth a year, with investment capital accounting for 35% of output and labour 65%. If labour grows by 1% a year and the capital stock stays flat, meaning productivity has to grow by about 1.35%. Now let labour shrink by 1% a year instead. The same 2% target needs about 2.65% productivity growth.

No doubt, that is a big gap. The economy has to get much more out of the same resources just to grow at the rate a growing workforce used to give it. The arithmetic concerns labour input, not simply the number of people, though working hours and participation matter as well.

Investment Capital can help. At least, firms can buy machinery, automate and invest in digital infrastructure. But that needs financing as well, and productivity gains don’t just arrive on their own. They depend on constantly changing technology, management, education and how well businesses actually use new tools.

So, a shrinking workforce doesn’t make growth impossible. It makes the sources of growth matter much more.

How Ageing changes saving and investment

People don’t save and spend the same way at every age. Working households for instance,  earn, build up assets and prepare for retirement. Later in life many draw those assets down. Economists have studied this for decades through lifecycle models.

Now if we apply it to a whole country. A society full of working-age savers can look very different in capital markets from one where a growing share of households has retired. Total savings, demand for financial assets and the supply of capital for investment can all shift.

To which direction one might ask. The truth is that the direction isn’t obvious, though. Because many retirees don’t sell their assets. Some keep saving, some hold wealth for emergencies or for their heirs, and pension funds invest over horizons much longer than any one person’s retirement. Pension systems, healthcare and income distribution differ a lot from country to country, and so does behaviour.

So will ageing population really push interest rates up or down? There’s no simple answer, and both arguments are plausible. On one hand, slower growth in population and labour reduces the need for new capital, and weaker investment demand could push real interest rates down. On the other hand, retirees drawing down savings, governments borrowing more for pensions and healthcare systems, and labour shortages pushing up wages could all push rates, and maybe inflation, up.

Which force wins depends on timing, saving habits, productivity, fiscal policy, migration and global capital flows. No ageing country stands alone, because its rates and asset prices are shaped by savers and investors in younger countries too. That’s why this is still an open research question, and why we’d be wary of anyone who turns a demographic chart straight into a forecast for bond yields. One must be very careful in this.

The pressure on public finances

Ageing population also changes the sums in government budgets. In many countries like the Europeans, pensions and healthcare are paid for by today’s workers, through contributions and taxes. As retirees grow in number compared with workers, spending on older people rises just as the tax base grows more slowly.

A useful indicator is the old-age dependency ratio. People aged 65+ divided by people aged 15–64, multiplied by 100. It is a demographic measure, not a literal count of pension recipients per taxpayer.

Yes, governments do have options. They could raise taxes, change the retirement age, adjust benefits, encourage immigration, get more people into work, or even borrow. Each one has costs. Higher taxes can weigh on spending and investment. More borrowing means higher interest bills. Changing retirement rules affects how much people work and save, and who bears the cost between generations.

Two countries that are ageing in population at the same pace can end up with very different credit risk, depending on their productivity, debts, pension rules and fiscal strength. Demographics can’t decide a country’s credit risk alone, but leaving them out leaves a big part of the fiscal picture unexplained.

The deeper problem is the models

At Schwarz Quant, this is what interests us most. Many models are estimated on historical data from decades when workforces were growing, labour markets were expanding and household saving followed a particular age pattern. They can describe the past very well. But a model that fits history isn’t guaranteed to hold up when the structure underneath it changes.

Take a forecasting model that links employment, household spending and output. If the working-age population starts shrinking, that link may weaken. More spending may come from pensions, savings and public transfers, and less from wages. A model linking interest rates to growth may miss what demographics do to saving, investment and fiscal policy.

The fact is, that the maths isn’t necessarily wrong. Because the relationships it describes may just not behave the way they did in the period used to estimate them. Economists call this structural change, or parameter instability. A model can get better and better at describing the past while getting worse at describing what comes next, and no amount of statistical sophistication fixes an assumption that no longer fits the economy.

Where demographic change creates investment opportunities

Ageing population isn’t purely bad news.

A smaller workforce raises the value of technologies that let fewer people produce more, such as automation, robotics and AI.

Demand for healthcare and for services used by older people will shift. Businesses that can do more with fewer workers may have an advantage over those whose model depends on a large labour pool. There may also be opportunities in private equity, infrastructure and corporate restructuring as businesses adapt their production processes, financing arrangements and long-term investment strategies.

Still, a growing industry and a good investment are not the same thing. A company in a booming market can still disappoint if it is overpriced, expensive to finance or unable to keep the gains from competitors. The same goes for countries. An ageing economy isn’t automatically a poor one. Productivity, investment and the quality of institutions may count for more than the demographic trend on its own. The opportunity lies in understanding how these forces interact and where financial markets may be mispricing their consequences.

Our perspective

At Schwarz Quant, we think population ageing deserves more attention in the assumptions behind macroeconomic forecasts, sovereign credit analysis, company valuations and long-term portfolio design. The question isn’t only which countries are ageing fastest. It’s what happens to the economic relationships people have relied on when the shape of the population changes.

Some of the questions we’re working on:

• How much productivity growth does an economy need to make up for a shrinking workforce?

• What happens to the cost of capital when saving and investment demand move in different directions?

• How should government debt be judged when age-related spending grows faster than the tax base?

• And perhaps most importantly, which companies and assets are positioned to benefit from these changes, and which are valued on assumptions that may prove increasingly difficult to sustain?

None of these can be answered by demographic projections alone. They depend on how population trends interact with productivity, capital, fiscal policy and markets, and they need alternative scenarios tested, not the assumption that the past will repeat. Migration, participation, longevity, fertility and technology can all change the result. Projections are useful, but they aren’t certain.

That’s where we think quantitative research earns its place – not in producing forecasts with ever more decimal places, but in showing how sensitive a forecast is to the assumptions underneath it. Ageing population won’t hit every economy, industry or asset class the same way, and our interest is in understanding those differences and identifying where they may create opportunities for disciplined capital allocation and attractive risk-adjusted returns.

When the population changes, the economy changes with it also. A model built around yesterday’s demographics may not be the right one for tomorrow’s decisions.

This commentary is for general information and research purposes only. It is not investment advice, an offer or a recommendation to buy or sell any financial instrument.

Sources and methodological notes

• Eurostat, Demography of Europe (2026 edition), population structure on 1 January 2025: https://ec.europa.eu/eurostat/cache/digpub/demography/

• Statistics Bureau of Japan, Respect for the Aged Day statistics (September 2025), share aged 65+: https://www.stat.go.jp/data/topics/topi1460.html

• National Bureau of Statistics of China, Statistical Communiqué on the 2025 National Economic and Social Development: https://www.stats.gov.cn/english/PressRelease/202602/t20260228_1962661.html

• China annual population change (2022–2025): annual National Bureau of Statistics releases. Figures in millions: −0.85, −2.08, −1.39, −3.39.

• Model notes: Cobb–Douglas with constant returns to scale; growth-accounting identities use continuous growth rates. Figure 2 uses illustrative parameters, not estimates for any particular economy. Population shares and counts are not equivalent to labour input.