In the first essay in this series on the demographic future of humanity, I explained how there is a high probability that the human population will peak between 2050 and 2060. Since I wrote the essay, two new facts have reinforced my prediction. First, Chinese authorities announced that there were 10.6 million births in mainland China in 2021 (vs. 12 million in 2020). Given that there were 10.1 million deaths, mainland China’s population grew by a tiny 480,000 inhabitants (before net migration, which was likely to be close to zero). And since births decreased at a roughly constant rate during the year, it is most likely that China’s population peaked in the early summer of 2021 at slightly above 1.4 billion, and that its population is already falling now.
Second, India now estimates that its fertility rate is 2.0 (1.6 in urban areas and 2.1 in rural areas), below replacement rate. The updated estimates predict that India’s population will peak around 2050 at 1.6 billion (my personal forecast is that it will peak earlier and at a lower level). In the second essay, I discussed how population decline will bring some major economic challenges that will jeopardize the maintenance of the welfare state and the growth of economic productivity.
In this final essay, I will outline some foreseeable consequences of these trends that concern matters fundamental to society, such as the family or housing.
The task, however, is a delicate one. As a society, we are venturing into uncharted territory. What ultimately happens will probably surprise us; therefore, the following paragraphs should be understood more as evidence-based speculations than detailed predictions.
Low Fertility and Aging Populations
But before we begin such speculations, it is important to point out that the fall in population will be accompanied by its aging due to increased life expectancy. Although fertility decline and an aging population are often conflated as a joint phenomenon, they are two separate issues. A society can suffer a population decline without aging (as happened during late antiquity in the West), while, conversely, we can have an aging population with stable fertility. However, fertility decline and aging are often mutually reinforcing. Allow me to explain.
In the undergraduate class I teach on the Political Economy of Early America, I always emphasize that grandparents were “invented” in the late eighteenth century in New England. I put the word “invented” in quotes because I am not referring to grandparents as a biological concept (since the dawn of humanity, we have all had grandparents), but as a key social category in family relations. The reason is simple: a combination of high per capita income (New England at the end of the eighteenth century was the richest region on the planet, and it also had relatively equal income distribution) with a climate unfavorable to the era’s most common infectious diseases (it was cold but not excessively so in winter, with moderate heat in summer) made it so that, for the first time in history, a high percentage of people survived long enough to be involved in their grandchildren’s lives. The idea of what a “grandparent” was and how they should interact with their grandchildren spread to the rest of the U.S. and Europe throughout the nineteenth and early twentieth centuries, as life expectancy also increased in those regions.
A simple way of showing how low life expectancy in earlier times prevented grandparenthood is to look at past European monarchs’ age of death. It is useful to note because, being at the top of the social pyramid, kings have always had access to the best doctors, food, and living conditions. And, with a few exceptions, European monarchs of the last five centuries were not burdened (mostly by choice) by an excessive workload or a stressful life.
Since I am originally from Spain, I will pick the Spanish monarchs, from the Catholic monarchs who created modern Spain to Alfonso XIII—the last king to die. Moreover, unlike other European monarchies, no Spanish monarch has ever been assassinated, so we have a fair assessment of “normal” life expectancy without violence. If we eliminated the kings of other countries who were killed (e.g., Charles I of England or Henry IV of France), the results for Spain and other European monarchies would be roughly the same.
Since men and women have different life expectancies, let’s focus on the age of Spanish kings at their death. Ferdinand the Catholic: 63 years; Philip I: 28 years; Charles I: 58 years; Philip II: 71 years; Philip III: 42 years; Philip IV: 60 years; Charles II: 38 years; Philip V: 62 years; Louis I: 17 years; Ferdinand VI: 46 years; Charles III: 72 years; Charles IV: 70 years; Ferdinand VII: 48 years; Amadeo I: 44 years; Alfonso XII: 27 years; and Alfonso XIII: 54 years. The queens’ ages at their deaths were: Isabella the Catholic, 53 years; Joanna I, 75 years; and Isabella II, 73 years. By comparison, Juan Carlos I, who has already enjoyed the advantages of modern medicine, is 84, a ripe old age, but by no means exceptional in advanced countries in 2022. Think of it this way: if you are 76 years old, you have already lived longer than any of the Spanish monarchs from the Catholic monarchs to Alfonso XIII.
In other words, reaching the age of 60 was already an achievement for the vast majority of the population (life expectancy was much lower, around 40 years, but this was weighed down by very high infant mortality). We should add that, in Western Europe, marriage (beyond the aristocratic elites) was relatively late from the late Middle Ages onwards, with grooms around the age of 25 or 26 and brides between 23 and 24 (this is called the European model of marriage).
That is, if a man married at 25 and the first child was born at 26 (the percentage of children born before marriage or less than nine months after marriage was small), the first grandchild could only arrive when the man was close to 52 (assuming the firstborn survived long enough to have a child, which usually did not happen). For a large majority of people, their lives were not long enough to see more than the first pair of grandchildren and to be in contact with them for, at most, a few years. Grandparents were distant figures, with whom people barely interacted, if they got to know them at all.
One Child, Many Adults
Today, when it is common to reach the age of 90, one can have a child at 30, a grandchild at 60, and live to see one’s first great-grandchildren. The consequences of this change in intergenerational relationships, especially in a low-fertility environment, are tremendous. Many children grow up surrounded by a high number of adults, but with very few children in their family environment. In China, for example, it is a cliché to speak of the “little emperors”: four grandparents, two parents, one child. The children enjoy a comfortable life with six adults around them and no “competitors.” They are given a monopoly on the emotional attention due to offspring. I still remember the question I was asked by a fellow PhD student in the United States who grew up in China under the one-child policy. Upon learning that I came from a large family, he asked, “How does it feel to have a sibling?”
The idea of having several siblings seemed so different from his life experience (and from all the people of his generation he had met in China) that the picture of my siblings he saw in my apartment fascinated him. Is it any wonder that today’s teenagers face more socialization problems than past teenagers, especially when they leave home? Or should we be surprised at the difficulty of creating intragenerational support networks?
But let us think of some more concrete consequences of this inverted pyramid of social relations. In many families, having four grandparents also means an inheritance of two homes, and two parents means an additional home. In other words, there is a clear expectation of a large transfer of wealth in the future. But, at the same time, such an inheritance may be delayed by many decades. What effects will these large but delayed inheritances have on the labor supply, savings, and investment of today’s young people? A dangerous combination of greedy expectations and endless waiting that we can call “Charles, Prince of Wales syndrome.”
At the same time, young people will rightly tell me that they find little consolation in having received this attention from adults or in thinking about future inheritances when all the social rungs are “occupied” by previous generations. Again, it is easiest to see this with an example. Think of firms of freelance professionals (law firms, consulting firms, auditing firms) or university teaching—careers that historically have provided opportunity (not perfectly, but not insignificantly) for social advancement. When lawyers or professors retired or died at the age of 60 or so, the doors were opened for the promotion of those who came behind them in the career ladder. Today, one constantly sees lawyers or professors at the age of 75 (in particular in countries like the United States where retirement is not mandatory) who are still at the peak of their careers. And, while society as a whole and these older professionals gain from enjoying many additional years of productivity, it is undeniable that this also causes “bottlenecks” among those who come after them.
If we add to this longevity the fact that positions in freelance professional firms and in university faculties will be reduced as the population falls—we need far fewer professors in a country where one million children are born each year than in a country where two million are born, no matter how much we increase the percentage of university students in each cohort—we can better understand the anxiety of many young people who wonder when they will have the opportunities that their parents and grandparents had. Is it any wonder, then, that many of these young people respond to this uncertainty about their future by delaying family formation and having far fewer children?
All this becomes much more complicated when, in addition, we realize that these changes will not be evenly distributed throughout the territory, even if we concentrate on a single country.
With current fertility and reasonable immigration assumptions, it is very likely that China will lose around 600 million inhabitants over the next decades. Internal Chinese politics will not allow the arrival of 600 million immigrants, in particular since the only possible “sources” of immigrants are either India or Africa, and most Chinese will not easily accept these immigrants (this prediction is not a reflection of my preferences about the optimal level of immigration across countries). The problem is that those 600 million inhabitants are not only going to disappear from Beijing or Shanghai—they are going to disappear from the rural and interior parts of the country.
These big cities will continue to enjoy two advantages that they share with other large cities around the world. The first is that of urban life, which generates much higher labor productivity gains than in past decades: a phenomenon that economists still do not understand very well, but which is extremely well-documented.
The second reason is that large cities have abundant supplies of leisure goods (restaurants, bars, theaters, sporting events), which are in great demand among the younger generations. Large cities enjoy a tremendous comparative advantage in supplying leisure goods since there aren’t any adequate internet substitutes for them: you can work remotely, but “virtual disco” is not the same as disco in person. Moreover, these leisure goods go hand in hand with having few children: it is much easier to go to a trendy restaurant when you don’t have four boys to take care of at home. Many of my students in college tell me that their dream is to live in Brooklyn or San Francisco. I have yet to meet a student who has told me that their dream is to live in the back country of Pennsylvania (a lovely region in terms of natural beauty).
Also, in countries like the United States., the south and the southwest are going to attract more and more people because, as they grow wealthy or age, people tend to prefer warmer climates. Again, many investment bankers or successful lawyers I know have told me that they are saving up to buy a house in Santa Barbara, but I have yet to meet anyone who is saving up to move to North Dakota (the state with some spectacular national parks that everyone should visit). Maybe there are a few people out there dreaming of North Dakota, but it’s not the norm. The consequences of this accelerated depopulation of many regions will be felt across the board: first, and most immediately, in the political arena.
Consequences of Falling Fertility
Let’s concentrate in the U.S. New York state’s electoral college votes peaked at 47 in 1932. In 2024, the votes will be 28. In the same period, Florida’s electoral college votes have gone from 7 in 1932 to 30 in 2024. Let us think about this for a second. In 1932, New York State’s electoral weight was nearly seven times Florida’s weight. In 2024, it will be smaller. A U.S. president with the electoral map of 2024, or more pointedly, the electoral map of 2040, will be a very different person from the president with the electoral map of 1932. I do not give a lot of weight to the probability that I will ever see again a president from New York state in my lifetime (although politics is always full of surprises!).
The second consequence is that many regions will begin to run into problems in providing basic public goods, such as universities or hospitals, which require certain minimum sizes to operate with reasonable efficiency. There were 11,537 births in Maine in 2020. Let us imagine that 60 percent of those born in Maine in 2020 go to college in 2038, and that the number of students leaving Maine to study in other states or abroad is equal to the number of students arriving from other states or abroad (both assumptions are optimistic, especially when it becomes easier to enter into the universities in large cities as a result of falling fertility). This indicates that in 2038, there will be around 6,900 incoming college students in Maine. How many of the local colleges and universities can survive with that incoming cohort? I do not envy the job of Maine education administrators in 2038.
But this phenomenon will happen all across the United States. Births have collapsed since 2008, putting U.S. higher education in a delicate position. The big crunch in incoming cohorts will not affect Harvard, Princeton, or Penn that much. With a smaller total population of prospective students, top universities will only need to begin accepting applicants slightly lower in quality. For example, instead of accepting only the top 4 percent of applications, they will accept the top 8 percent of a smaller pool. The crunch will not affect Penn State or UC Santa Barbara that much either. The equivalent of some of their best students will go in 2038 to Ivies, but they can turn around and pick among the next best applications, the ones they are rejecting now just below the waitlist.
This ladder will move down the hierarchy of college selectivity until we get to the least selective level, where a lot of institutions will need to close or dramatically restructure. And no, bringing foreign students will not fix it. Recall my opening paragraph about China and India: because of the fast drop in fertility, it will be much easier for a Chinese student to get into Peking University in 2038 than in 2020. Thus, why bother coming to a second-rate U.S. institution?
Finally, we will see a radical change in housing prices. Let us return to the case of Maine or upstate New York. Who is going to buy all the empty houses left by a reduction in population? Some homes will disappear (demolitions or ruins in smaller towns, and the bad apartments built in the urban boom of the 1950s and 1960s will be abandoned); others will be transformed into second homes; and, finally, with smaller families, the average number of people per dwelling will fall. But, even discounting these effects, in 2040 or 2050 we will find ourselves with thousands of dwellings that nobody will want, which will push housing prices to the bottom.
Again, the biggest stress will come not so much from falling house prices, but the unevenness in price changes. Condo prices in Manhattan and houses in Santa Barbara will not fall unless there is a dramatic change in work and leisure patterns, which cannot be ruled out, but which I think is unlikely.
If current trends continue, in 2040 an apartment in Manhattan will be much more attractive than in 2021. But a house in the middle of upstate New York is going to be much cheaper. Families with real estate assets in the first set of locations will enjoy significant capital gains, and families with real estate assets in the second set of locations will suffer.
If this scenario worries you, and you want to propose a solution, ask yourself if the solution will work with a population that’s falling quickly (China) or stagnant (the United States) and whether it’s compatible with young people’s lifestyle preferences.
I shall conclude here. In these three essays I have written almost ten thousand words on humanity’s demographic future. Explaining all the ideas in these essays in detail, or presenting many more figures and justifying the empirical evidence, would easily take me a whole 250-page book. We will leave that for another time. At the same time, I have repeatedly tried to emphasize that society is a complex system that changes in unexpected ways. It may well be that all that I have said will come to nothing in a decade. But it is a fallacy to argue that because the future is uncertain one shouldn’t worry about current trends or do anything about them. Precisely because the future is uncertain, we should be even more careful and consider all scenarios. Unfortunately, most of us are not thinking in any detail about our demographic future.
This article is adapted from a version originally published in November 2021 at El Confidencial, a leading Spanish digital newspaper. We are grateful to Professor Jesús Fernández-Villaverde for his permission to publish it in English here, and to Thomas Howes for his translation.