Every three years the European Commission forecasts the cost of ageing. They recently published an update. The logic is interesting, the assumptions sometimes surprising and the results predictably depressing.

The Commission is interested in what these costs will be by 2070. They are particularly interested in the share of total GDP taken by the costs of ageing. In 2022 the EU spent nearly a quarter of its GDP on ageing. To put that into context. Expenditure on defense in those same countries was 2.5%.

The Logic

Forecasting such a number is an interesting intellectual exercise. The total ageing costs in 2022 represented 24.4% of GDP. They divide the costs into four “buckets”. The biggest is pensions. Since they are interested in the Government budgets these are the state pensions. These represented 11.6% of EU GDP in 2022. The second is Health Care at 6.9% of GDP. An ageing population increases the burden on medical care. The third is the cost of long-term care other than medical care (1.7%). From these they can deduct the reduction in educational costs. Society is ageing not just because we are living longer. There are fewer children. Education costs were 4.4% of GDP in 2022 but falling.

They are interested in the long-term sustainability of public finances. They need a GDP forecast as well. They can get a population forecast from their statistical bureau. They then need assumptions about the labour participation within the working age population. Finally, they need to know future productivity gains. Population decline and ageing changes the trajectory of GDP growth. The impact depends very much on the productivity assumption.

The Assumption Surprises

Having written a lot about fertility forecasts, for me, these are the most problematical. EUROSTAT the statistical bureau, follows the UN method. It assumes that there will be an inversion. At some point prosperity will grow sufficiently for children to be an attractive “investment”. They are forecasting that by 2070 the fertility rate will be higher than in 2022 in every EU country. They are not forecasting above replacement level in any country. They are however expecting a slowing in the rate of decline, a plateau and then a mild increase. Other forecasts are based on a range of variables. These include the level of female education and the availability of contraception. They even add the level of urbanization. These forecasts see no let up in the decline.

I have written about the increase in people working at older ages. The Commission does include a 3% increase in participation in the workforce. This is not enough to compensate for the decline in the working age population. It does recognize the changes in pensionable age. But it includes nothing for working beyond retirement age.

Across the whole of the EU, Governments are managing the impact of the dependency ratio. This was 29% in 2010. It rose to 36% in 2022. The forecast is a rise to 59% by 2070. The bulk of this will happen before 2045. In particular, pensions as a share of GDP are forecast to only rise from 11.4% to 12.7% but then fallback to 11.8%. Retirement dates are assumed to extend. At the same time payouts relative to labour rate inflation are assumed to fall. Only one country plans to maintain the ratio of pensions to labour rates. The others plan to cut relative pensions. How politically robust these assumptions are is a different question. There was a huge push back against the French increase in retirement age.

Some assumptions are less surprising. Long term care costs are forecast to increase from 1.7% of GDP in 2022 to 2.5% in 2070. There are increasing numbers of people above 80. In the absence of a ramp up in defense spending this means countries will spend as much on long term care as defense. Education costs fall over the period. With lower fertility forecasts they would fall further. However, lower fertility means there would be a smaller working age population. This would more than offset any savings.

The Variation Across Countries

The forecast is for the “costs of ageing” in the EU to increase from 24.4% to only 25.6% in 2070. Not an apparent large increase. But this must be set in the context of all the other demands on Governments. But, there are huge variations across countries.

In 2022 there were already large differences. France and Norway both were spending 30% of total GDP on these costs. Romania by comparison was spending only half of this amount. The Commission is forecasting no increase in the share of GDP taken in Greece, Italy, France, Portugal, Croatia and Latvia. But, there are large increases for thirteen countries. These include Italy, Spain and Belgium. The differences depend on their histories. On the position taken in each country over many years on the provision of these services to their population.