Economists have long warned that the aging of the population would make our lives harder on two fronts. The first is growth. For decades, Spain benefited from the so-called demographic dividend: a growing workforce and many workers for each retiree, a tailwind that propelled the economy with little effort. That dividend has been exhausted and, in many respects, has turned negative. With fewer workers relative to the total population and a workforce that is aging, the growth potential is reduced. And if you don’t grow, you can’t live better.
The second front is fiscal. An aging society needs more pensions, more healthcare, more elderly care. Spending rises structurally, not cyclically. And it does so precisely when the base that finances that spending—the employment, economic activity, and revenue—grows more slowly. The result is a narrowing of fiscal space that makes it increasingly difficult to pay for the things we want to pay for.
“An aging society needs more pensions, more healthcare, more elderly care. Spending rises structurally, not cyclically”
What matters is that all of this is not a future threat. It has arrived quietly and is already here. The baby-boom generation is entering retirement. The numbers we saw in the models are now appearing in the public accounts. And as is often the case with problems that arrive slowly, when they finally become visible there is very little room to act without cost.
According to the latest demographic projections from INE, published in June 2024 with a horizon to 2074, Spain will move from 48.6 million inhabitants today to 54.9 million in 2050 and 54.6 in 2074: an increase of only 12% over fifty years. A figure that, at first glance, might seem reassuring. But what matters is not how many we will be, but how we will be distributed by age. The total dependency rate will rise from the current 53% to 73% in 2050. And the elderly dependency rate —the ratio of those over 64 to the working-age population—, which matters for pensions, will go from 30% to 50%: from 1 person older than 64 for every 3.3 workers today to 1 for every 2 in 2050.
The most striking aspect is that this intense aging process will occur despite projections assuming net immigration of 17.1 million over the period. Immigration is essential —without it the deterioration would be much more severe—, but it is not enough to reverse the trend. This is explained by two forces acting simultaneously. Fertility barely recovers: Spain recorded in 2022 a historic low of 1.16 children per woman, and projections only raise it to 1.24 in 2038, far from the 2.1 needed to replenish the population. At the same time, : life expectancy at birth will move from 80 for men and 86 for women today to 86 and 90 years, respectively, in 2073. By the age of 67—the legal retirement age—a man who retires today will collect his pension for an average of 17 years; in 2073, it will be 21. The age structure, not the size of the population, is what changes everything.
The pension system: the best invention of the 20th century that needs to adapt
Aging is going to press on all fronts of the welfare state, but in this article I will focus on pensions.
The pay-as-you-go pension system —the one where today’s workers fund the pensions of today’s retirees— is probably the best social policy invention of the 20th century. In it, workers accept to give up part of their wages not because they immediately get something back, but because they trust that future workers will do the same for them when their time comes. It is, in essence, an intergenerational pact. This trust has pulled millions of older people out of poverty and given security to generations entire.
“Workers accept to cede part of their salary not because they receive something in return immediately, but because they trust that future workers will do the same for them”
But every contract has implicit conditions. And the demographic conditions on which this pact was built have changed radically. Without dismantling the system, we must recognize that it was designed for a world that no longer exists, and adapting it to longevity is not an attack on the intergenerational pact, but, precisely, the condition for that pact to remain possible.
But to reform something one must start by telling the truth. And here Spain has a serious transparency problem because the official figures of the system convey an image that does not match reality.
The deficit published by Social Security — €7,352 million, 0.4% of GDP — seems manageable. However, that figure only captures the system’s accounting balance, not its real financial position. When Social Security and the regime of Civil Servants are consolidated, and the contributory component —the part funded by contributions and supposed to sustain itself— is separated, the real deficit rises to €61,881 million, 3.7% of GDP, according to a recent report by the Santalucía Institute authored by Miguel Ángel García Díaz. The gap between what is communicated and what actually happens is almost €55 billion.
This mismatch is not new, but it has worsened. In the same study it is shown that, between 2018 and 2025, contributory revenues grew by 50%. Expenditure grew even faster: 54%. The contributory deficit has nearly doubled. To cover it, the State has had to inject growing amounts: €33.953 million in 2018, versus €58.395 million in 2025. Since 2010, the State’s cumulative contribution to the system exceeds half a trillion euros, the equivalent of 50% of all public debt issued by Spain in that period.
The most worrying thing is the context in which all this has occurred. Spain’s economy experienced years of robust job growth: 2.8 million more contributors between 2018 and 2025. The conditions could hardly have been more favorable. And yet, the hole has only grown. The reason is simple, though troubling: the bulk of the baby-boom generation has not yet reached retirement.
What lies ahead
The arithmetic of demographics is decisive, and its consequences for pension spending are predictable. When AIReF and FEDEA project what lies ahead, there is little room for complacency. AIReF puts pension spending at 16.1% of GDP in 2050; FEDEA raises it to 17.1%. Today we are around 13%. We are talking about three to four percentage points of GDP more for pensions in barely a quarter of a century.
And this will occur in a context where the economy will grow less, precisely because there will be fewer workers in proportion to the population. The gap that already exists — and, as we have seen, amounts to 3.7% of GDP when measured rigorously — will widen much more when the baby-boom generation begins retiring in large numbers. This is one of the direct consequences of the demographic reality we face.
The political economy of pensions
In this context, we must ask: can Spain finance a pension spending increase of this magnitude in an environment where the number of people receiving benefits grows and the number contributing to the public coffers shrinks? The honest answer is yes, but with a cost that is rarely made explicit. Every euro of fiscal margin allocated to pensions is a euro not available for health, education, housing, fighting child poverty, or infrastructure investment. Economists call this the opportunity cost. It’s a technical term for something simple: spend more on one thing means spend less on others.
“The conditions could hardly have been more favorable. And yet the hole has only grown”
Here is where political demography enters. The median voter in Spain today is 49 years old. In 2050, according to INE projections, they will be between 57 and 59. Those over 50 already represent the electorate’s majority, vote more than the young, and have a more homogeneous agenda. For any party hoping to win elections, the temptation is obvious: pensions take priority, at any cost. No conspiracy is needed to explain it. It’s simply the logic of suffrage in an aging society.
But something being electorally profitable does not mean it is economically sound. And here it is useful to make a distinction that political debate often overlooks. A portion of the pension spending increase is perfectly justified: if the old-age dependency rate almost doubles, it is logical that there are more retirees and that the system costs more. That higher spending is the natural price of aging, and there is no reason not to pay it. The problem lies with the other part: the increase in spending that arises because we refuse to adapt the system to the new longevity. When someone retires today at 67, they collect a pension for an average of 17 years; in 2073, it will be more than 22, whereas in 2000 it was less than 13 years. The system was not designed for that, and continuing to ignore it has a price paid by others: resources that do not go to reducing child poverty, improving education, building affordable housing, or preparing the economy for the future.
Economically, what must be done is obvious. Politically, it is very difficult.
A possible reform
Before talking about solutions, it is worth recalling the starting point. The pension reform of 2021-2023 has the dubious honor of being the only reform of the 21st century in an industrialized country that has resulted in an increase in the system’s generosity. While all our European partners had decades of adjusting their systems to the new demographics, Spain had also moved in that direction: the reform of 2011 under Zapatero raised the retirement age from 65 to 67, it was politically costly, but it was moving in the right direction. However, the latest reform, instead of continuing that path, reversed the course. Precisely the opposite of what the rest of the industrialized world did.
“People over 50 already account for the majority of the electorate, vote more than the young, and have a more homogeneous agenda”
Does a politically viable reform exist then? Yes, but only if the adjustment is distributed across all generations. The following outlines a set of measures conceived to distribute the burden —among current pensioners, active workers and future generations— and minimize the political cost of an inevitable reform.
1. Reduce the generosity of the system
In Spain, the replacement rate exceeds 70%, one of the highest in the world. Our contributory system, with its traditional Bismarckian design, was built on the principle of income replacement, i.e., to resemble the retiree’s former wage. With a life expectancy of about 78 years that was workable. Yet, with an expectancy approaching 90, maintaining a 70% replacement rate while paying pensions for 25 or 30 years instead of 15 becomes unsustainable, unless someone is willing to pay the bill.
“Does there exist, then, a politically viable reform? Yes, but only if we distribute the adjustment among all generations”
Reducing generosity is inevitable; therefore, the question is how to do it fairly. There are two levers. The first, to count the entire working life to calculate the pension, which is already the Spanish tradition. The second, to restore the sustainability factor: an automatic mechanism that adjusts the initial pension when life expectancy at retirement increases. The logic is common sense: if two workers have contributed the same over their working life, it is fair that they receive the same total during retirement. Since someone who retires later will live longer, they will receive less each year. It’s not a cut made arbitrarily; it is an automatic adjustment to longevity.
This mechanism was approved in the 2013 reform of President Rajoy and repealed in the last reform without ever having been applied. And it’s worth remembering: the social partners have an explicit commitment to approve a equivalent mechanism before 2027.
Both levers together —lifetime contribution base and sustainability factor— effectively convert de facto a defined-benefit system into a defined-contribution one. And that, almost inadvertently, lays the groundwork for the natural step: a notional accounts system, in which every worker accumulates over their career a virtual capital that determines their pension. It would be the optimal solution. But that is another conversation.
2. Delay retirement in a flexible way
aging will reduce the number of workers as part of the population, even assuming strong immigration. A portion of that loss can be offset by leveraging senior talent: there are millions of experienced workers who leave the labor market earlier than they would like, or earlier than the economy needs.
“The alternative is a gradual exit from the labor market: work part-time while receiving a portion of the pension, making a gradual transition to full retirement”
But delaying retirement cannot be done with the current all-or-nothing model: working full-time until the last day and then full retirement the next day. That does not work for workers or for firms. The alternative is a gradual exit from the labor market: part-time work while collecting a portion of the pension, making a progressive transition to full retirement. In Spain there already exist tools for this —active retirement and partial retirement—, but they are barely used. The challenge is to make them attractive for both workers and the companies that retain or hire senior talent.
Moreover, this flexibility must be accompanied by fairness. Postponing retirement cannot be a one-size-fits-all. A construction worker, a miner, a police officer or a firefighter are not in the same situation as an economist or a lawyer. For physically demanding or health-impaired occupations, and for workers in poorer health, retirement must remain possible earlier and under more generous terms.
3. Promote the Dependency Law
The third measure is to push the Dependency Law forward. And here there is an argument that goes beyond social justice: it is an efficiency argument.
Many retirees save too much during their retirement out of fear of becoming dependent and unable to pay for care. It is a preventive saving against a risk that may never materialize. Because when we analyze how people die, the reality is more varied than fear suggests. Some spend many years in a state of heavy dependency, some require little care, and some require none. But since no one knows in advance which group they will be in, public insurance would allow sharing risk among many.
“There are people who spend many years in a state of great dependency, there are those who need little care and there are those who need none”
Seen this way, a robust public dependency coverage isn’t that expensive. The share of people who remain in a high-dependency state for more than a year is small. What makes dependency devastating is not its probability but its cost when it occurs. It is for this reason that collective insurance is more efficient than individual saving.
The proposal, therefore, is not to spend less, but to spend better. Slightly reduce pension generosity to offer real protection against dependency. A retiree with a pension slightly lower but with real long-term care coverage may be objectively better than one with a higher pension but exposed alone to that risk.
That is what we mean when we talk about modernizing the contract between generations. Not cutting the welfare state. Redesign it to better protect against the risks that truly matter.
4. Promote the second pillar: supplementary savings
To understand why it matters so much, it helps to discuss demographic dividends.
The first has ended. In the past, the rise of the working-age population relative to dependents provided a growth wind: more workers, more production, more resources to finance the welfare state. That dividend has run out and the trend is moving in the opposite direction. At the same time, aging brings an opportunity: if people expect to live longer, they need to save more to fund a longer retirement. That savings, well invested, can generate the second demographic dividend, producing more capital, higher productivity, and more growth.
To capture it, we must push the second pillar: employer pension plans that complement the public pension without substituting it. The first pillar—the public PAYG pension—will always be the base. But it cannot be the only one.
“The proposal, therefore, is not to spend less, but to spend better. Reducing somewhat the generosity of pensions to offer real protection against dependency”
The problem of voluntary saving is well known: few people save on their own, even though they know they should. The solution comes from behavioral economics: auto-enrollment. Workers are automatically enrolled in a pension plan, but they can opt out, since it is not mandatory. What changes is the starting point by moving saving from an active decision to the default option. The change is not minor. Evidence from countries that have implemented this shows that more than 90% of workers stay in the system. When saving is normal, people save.
This saving, channeled through long-term pension funds, is invested in infrastructure, companies, innovation, energy, housing or technology. It increases the economy’s capital stock, boosts productivity and drives growth. And, moreover, allows offsetting the decline in the generosity of the public pension: the worker receives a bit less from the State, but accumulates personal savings that supplement their retirement.
It is, incidentally, exactly what the Draghi and Letta reports recommend to strengthen European competitiveness: mobilize private savings of citizens and invest them productively in the long term. Spain has an open task here.
If we do nothing, we already know what happens: all the fiscal margin in the coming decades will go to pensions. The demographic arithmetic leaves no alternative, so it will happen without anyone’s explicit decision.
Reforming the pension system is not an attack on the elderly. It is a matter of intergenerational justice. The contract between generations that sustains the PAYG system is legitimate only if it is balanced: if those who work today can trust that the system will be there when they retire, and if those who retire today do not jeopardize the opportunities of those who come after. That balance has been broken and there are reasonable ways to restore it, as we have tried to illustrate.
One final warning. Every year that passes without reform makes the problem bigger and the solutions more costly. That cost, as always, will not be borne by those who hold political power to prevent it. It will be borne by the youngest, who are precisely those with the least voice in a system where the median voter ages year by year. The political economy of the problem is also the most urgent reason to resolve it.
In collaboration with the “la Caixa” Foundation
