September 26, 2019 Reading Time: 7 minutes

Among the developed nations, fertility rates have been declining since the first quarter of the 19th century, even if the trend has not been linear. This factor, combined with increases in life expectancy, has created one of the greatest challenges to society today. As the working-age population shrinks, the ranks of the aged swell. 

The 2019 United Nations population survey predicts that the world population will rise from 7.7 billion today to 11.2 billion by 2100. That growth will be driven mainly by the nations of Africa, whose population is expected to expand from 1.34 billion to 4.28 billion by the end of the century. By contrast, 55 countries will see their populations decline, including 26 where the number will decline by more than 10 percent. UN statisticians predict that, globally, the fertility ratio will decline from 2.5 to 2.2, while longevity will rise from 72.6 years to 77.1 years by midcentury. 

Two hundred years is a relatively short sample period, but it is probably safe to predict that rich, successful societies, especially ones where women are well-educated and have reproductive freedom, choose fertility rates that imply gradual population decline.

As any financial advisor will tell you, as we age, our economic needs change too. Upon joining the labor force, we plan to start our own families. Some of us save; others continue to spend their newly minted wealth, only to discover that saving would have been prudent. The rearing of children is an expensive pastime. 

Once our children have grown and flown, we switch our economic focus to the task of saving enough to retire. If we are healthy, this is easier than if we must pay for medical care, but most of us are confronted by similar lifestyle considerations. When we finally retire, as the economist Franco Modigliani discovered as long ago as 1954, we continue to spend, but, unlike our younger selves, we consume far less conspicuously.

At the national level, this aging process can be seen everywhere, not least in the level of inflation and interest rates. In this scenario, central bank policies, aimed at stimulating consumption by artificially lowering interest rates, are akin to King Canute’s attempt to turn back the tide. An aging nation will consume less. The retired must live on their savings; if the rate of interest on those savings falls, they will be forced to consume less. For many, the calculations they made about the capital needed for retirement have proved inadequate. As a wise man once said, “It’s not that I saved too little; it’s simply that I never expected to live this long.”

For the governments of aging countries, charged with the provision of benefits for the retired, the problems are severe. They too have miscalculated. State pensions are predominantly paid out of current taxation; the system is unfunded. The robbing of Peter to pay Paul only works if the ratio between those who work and pay taxes and those who do not is sufficient to meet the needs of the latter. 

Nowhere are these trends more evident than in Japan. It is a country in the vanguard of demographic decline. Its response to the indefatigable march of time has been tentative, but there is hope, as always, in human ingenuity. Our ability to automate tasks once performed by humans, across agriculture, industry, services, and other sectors, enables us to do more with less. 

Taken to its logical conclusion, if automation is successful, it obviates the need for a growing workforce to meet the welfare needs of its aging population. Since man first picked up a stick to plow a furrow, we have been on a path to improve productivity by means of innovation: this trend is far from new.

Much has been written about the fate of workers, as artificial intelligence (AI) usurps the role of humans in innumerable professions. A 2013 study by Oxford University caused consternation, suggesting that, in the coming decades, 47 percent of total U.S. employment was at risk of automation. In a global study, the International Labour Organization warned that this was not just a developed-country problem. It predicts that 56 percent of the workforce of Cambodia, Indonesia, the Philippines, Thailand, and Vietnam is at risk of technological substitution within 20 years. It transpires that Asian robot sales increased 19 percent in 2015; the process has already begun. 

The table below indicates which U.S. jobs are most at risk, but a similar picture is evident in other developed and developing countries. The inflection point will be determined by average earning per capita.

Source: McKinsey Global Institute, U.S. Bureau of Labor Statistics
Source: McKinsey Global Institute, U.S. Bureau of Labor Statistics

McKinsey Global Institute, which compiled this research, concludes that roughly 60 percent of occupations are at risk of 30 percent automation, just based on the development of existing technologies. There is a ray of hope, however: the authors note that the benefits of that same automation could boost global productivity growth by 0.8 to 1.4 percent annually. 

If you are still inclined to view automation as a risk rather than an opportunity, then you may be more heartened by a 2016 OECD study of 21 of its member nations. The OECD’s researchers are less optimistic about the degree to which all the stages of a bundled set of tasks (which is typical of the majority of business processes) can be automated. It believes only 9 percent of jobs are truly at risk from automation. 

The table below, taken from the January 2018 MIT Technology Review, highlights the difficulty with making predictions about the future of employment and the impact of technology.

Source: MIT Technology Review

The article concludes: 

In short, although these predictions are made by dozens of global experts in economics and technology, no one seems to be on the same page. There is really only one meaningful conclusion: we have no idea how many jobs will actually be lost to the march of technological progress.

Before you throw your hands up in despair, it is worth considering that, if automation is inevitable, a fertility rate of less than two could be a boon rather than a curse. In fact, for nations with higher fertility rates, including many on the African subcontinent, the challenge of automation is greater than for those with aging populations.

The Graying of Japan

In a June 2018 publication, the IMF’s Finance and Development suggest that Japan’s combination of artificial intelligence and robotics may be the answer to its labor-shortage problem. Already a quarter of Japan’s population is over 65 and the number of people of working age (15 to 64 years) is projected to fall by 24 million by 2050.

Source: United Nations
Source: United Nations

Japan has been facing the demographic challenge of limited resources — both of labor and of natural resources — for many years. Here is how it ranks in robots per capita.

Source: International Federation of Robotics
Source: International Federation of Robotics

Japan remains the largest exporter of robots, although, since 2009, it has slipped down the world ranking by usage. The country’s most productive industrial sectors are automotive and electronics, yet 75 percent of Japanese GDP is still generated by services. It is here that AI will be deployed most effectively, engendering substantial productivity gains. These gains are long overdue; productivity in Japanese manufacturing has tripled since 1970, while in nonmanufacturing the gain has been a meager 25 percent.

The subsectors that should see dramatic gains include transportation, communications, health care, finance, retail services, and storage. The Japanese have even introduced robot chefs and hotel staff. The next phase will see robots replacing humans in schools, hospitals, nursing homes, airports, railway stations, and many other service industries, where simple, repetitive tasks can be more efficiently and cost-effectively performed by machines. 

The migration to machines is not occurring without issues. As labor shortages become more severe, Japan has seen the quality of human services decline. Its labor force is bifurcating into a minority of workers with the training and education required to benefit from robotic productivity. Among the losers are female workers who have only recently joined the labor force and those who have taken on temporary roles. 

The ultimate technological solution to the twin headwinds of an aging population and shrinking workforce will be better service at lower cost. These are the good deflationary forces of human progress. In the long run, Japanese consumers will end up paying less and getting more. 

For a more futuristic take on how automation is solving the problems of aging and labor shortages in Japan, this article from Towards Data Science (“Robots, Aging Population and Pets”) is instructive. Here is a taste: 

In a nursing home in Natori, Miyagi Prefecture, a Humanoid robot asks an elderly nursing resident in her 80s, “Grandma, what color flowers do you like?”. The woman responds, “Pink cherry blossoms”.

This humanoid is the first of its kind to be used in a Japanese retirement home. It is developed by Hiroshi Ishiguro and his team of engineers from Osaka University. Since the use of this humanoid, nursing home staffers report less irritation, improved cognitive functions and improved mental health conditions from residents in this nursing home.

A Sapporo-based company Safety Net linkage is developing smartphone applications in conjunction with this humanoid to help to track down people with dementia. These applications will allow families and nursing home facilities to keep an eye on elderly residents in cases of missing persons.

Easing the Adjustment

Advances in technology have always created societal challenges. For developed-nation governments, labor-market reforms are key. A focus on supporting workers rather than creating jobs is at the heart of the philosophical change required. Improving the quality of technical and vocational education and training is critical. For more than a decade, internet-based video technology has been capable of delivering both short- and long-form education. In theory, the cost of training has been vastly reduced. As governments’ tax base shrinks, their budgets will decrease. They will need to adopt these productivity-enhancing technologies to deliver educational services at lower cost.

There will inevitably be calls for government subsidies and tax incentives to encourage corporate investment in training and skills, focused on areas where humans have a comparative advantage over machines. There are worse things to spend our taxes on. The cutting of red tape would be better. Greater flexibility in labor regulations will be needed to compensate for the relative increase in the cost of skilled workers. Corporations and their employees can both benefit from flexible contracts and continuous professional development.


Declining fertility rates, increasing longevity, and technological progress are trends with strong momentum. If embraced rather than ignored — or, worse still, impeded — productivity can be boosted while the quality of life and of employment can be improved. Demographic decline is inevitable; if approached as an opportunity, it can be a blessing rather than a curse.

Colin Lloyd


Colin is a macroeconomic commentator, writer and presenter, based in London, England. He has worked for asset managers in commodities, money markets, capital markets, equities and foreign exchange since the early 1980’s and writes In the Long Run. He is a contributor to several free-market publications including The Cobden Centre and was a 2017 runner-up for the 2017 Richard Koch Breakthrough Prize awarded by The Institute of Economic Affairs.

Get notified of new articles from Colin Lloyd and AIER.