The world's changing demographics will have a far-reaching impact on our economy.
Context, as we know, can be very important in economics and in investing. Some of the most successful investors of our time might have been unknown humble laborers if they had instead been born in a poor country far away or born in this country at a less propitious time. Context is made of several components including, among others, political risk, the rate of innovation, fiscal and monetary policy, and of course demographics. Some or all of these components can remain unchanged for years or even decades, which may lead a majority of economists and investors to mistakenly view them as permanently fixed. Yet each inevitably comes to an inflection point which destabilizes economic or investment projections built on assumptions derived from the old paradigm.
In the case of demographics, they have acted for decades as sustained tailwinds for the US and global economies. The main drivers of these tailwinds were 1) the rise of the baby boomers and 2) the subsequent decline in the birth rate in North America, Asia and Europe, which resulted in a fall of the dependency ratio (number of dependents per working adult). Because these tailwinds have now largely died down (except in India and other parts of Asia), demographics can no longer be seen as a fixed component of the economic or investment context. What was true for decades is no longer true because we have recently passed an inflection point in demographics.
Going forward, changes in the populations of North America, Europe, Asia and Sub-Saharan Africa will likely undermine economic projections derived from the habits and assumptions of an obsolete context. Unlike the past few decades, demographics should now be considered as a moving variable which may be supportive or adverse to one's economic or investment thesis. (In my view, every investment portfolio should be subjected to a 'demographic audit' which incorporates the impending changes).
Some demographic megatrends were quantified in 2010 by a United Nations report, World Population Prospects (pdf), and are summarized in the table below. Forecasting is a difficult endeavor and the UN tries to mitigate the uncertainty by creating four different scenarios, or 'variants', for population projections: constant-fertility, high, medium and low. The constant-fertility variant assumes that the fertility rate (the number of children per woman) in each country and region of the world remains at the same level as it was in 2005-10. This variant shows a shocking increase in the world population to levels which are probably unmanageable, from 7 billion today to 11 billion in 2050 to 27 billion in 2100. It is highly unlikely therefore that fertility rates will remain unchanged. They are today exceedingly high in Sub-Saharan Africa and exceedingly low in Russia, Germany and Japan. The three other variants all result in lower population counts for 2050 and 2100. I use the medium variant below and all figures are UN estimates, not my own (for more detail on fertility assumptions of the four variants, see pages 27-35 of the UN report (pdf)).
The key points are as follows:
The population in each of the more advanced economies of North America, Europe and Oceania will either grow slowly, stagnate or fall precipitously. In the US and Canada, it will grow slowly. In a large majority of European countries, it will stagnate or shrink moderately. And in Russia, Japan and Italy, it will fall or fall precipitously.
Total Fertility (children per woman)
Europe faces a significant demographic challenge. It is in its causes and chronology similar to the challenge we face in the United States but it is more severe because Europe has a lower fertility rate. How do you keep the economy growing when the size of the population and its age distribution are no longer working in the direction of growth? It can be done but it is certainly more difficult. And how do you maintain Europe's cherished social programs when the number of workers stagnates or declines and the number of retirees increases?
Europe's population is expected to fall from 738 million in 2010, to 719m in 2050 to 675m in 2100. Because of some growth in Ireland, France and the United Kingdom, the population of Northern Europe would maintain itself or grow modestly. But it will decline in Southern Europe in large part because of Italy, Portugal and Serbia. On medium variant estimates, the number of Italians would shrink from 61m in 2010 to 59m in 2050 to 56m in 2100. Germans would also be fewer, from 82m to 75m to 70m. Eastern Europe (including Russia) would shrink from 295m to 257m to 222m, with every state except the Czech Republic, Hungary and Slovakia losing 20 to 30% of its population count by 2100.
The UN report projects that, at constant-fertility rates, the population of Russia would fall from 142 million in 2010 to 114m in 2050 to 67m in 2100. The more probable medium variant predicts a Russian population of 126m in 2050 and 111m in 2100, still a decline of 31m by the end of the century. Russia suffers from a low fertility rate and a low life expectancy, two factors which are likely to improve in coming years.
Africa presents the opposite profile with the population of Sub-Saharan Africa continuing to grow rapidly. In 2010, the fertility rate in Africa was 4.37 children per woman (5.43 in Nigeria) compared to 1.59 in Europe and 2.08 in the United States. Although fertility rates are expected to fall dramatically, the population of Africa will first grow from 1.02 billion in 2010 to 2.19b in 2050 to 3.57b in 2100 (again, using the UN's median variant). Of these numbers, Sub-Saharan Africa which has 856 million people today will total 1.96b and 3.36b. Given that the world population is expected to grow from 6.9 billion in 2010 to 9.3b in 2050 to 10.1b in 2100, it is easy to see that a huge part of this growth, 77%, will be coming from Sub-Saharan Africa.
If these numbers are surprising, consider that the world fertility rate stands today at 2.45 children per woman but that it is 4.78 in Sub-Saharan Africa. And while the Sub-Saharan rate is expected to decline to 2.85 by 2050 and 2.14 by 2100, it will not decline fast enough to avert the incoming boom. Indeed, the numbers above already assume such a decline. A fall in the fertility rate usually follows an improvement in health care and a fall in the mortality rate. There is excellent news on this front. As recently reported by the Economist, "16 of the 20 African countries which have had detailed surveys of living conditions since 2005 reported falls in their child-mortality rates". The World Bank calls this "a tremendous success story that has only barely been recognised".
China, India and Japan
China was the most populous nation in 2010 with 1.35 billion people but it will be overtaken by India around 2020 when both will have 1.39 billion. The impact of China's one-child policy means that its population count will fade to 1.3b by 2050 and 952 million by 2100, while India continues to grow to 1.69b by 2050 before it also fades to 1.55b in 2100. The assumptions built into these numbers are not extreme. Starting at 1.56 in 2010 (well below the world average), China's fertility rate would rise to 1.81 by 2050 and 2.01 by 2100 . India's now 2.54 (slightly higher than the world average) would fall to 1.84 by 2050 and tick up to 1.88 by 2100.
Like Europe, Japan's population will fall by a large percentage. Japan's fertility rate of 1.42 was in 2010 one of the lowest in the world, but the UN is expecting it to recover to 2.04 by the end of the century. This will not be soon enough to avert a precipitous decline from 126m Japanese in 2010 to 109m in 2050 to 91m in 2100.
The UN expects the population of North America to grow from 344 million in 2010 to 447m in 2050 to 526m in 2100 with nearly all of this growth taking place in the United States (respectively at 310m, 403m, 478m). My own estimate of the US population, published in America Heading for Zero Population Growth?, is lower. Without new immigration, I found that the US population would not grow at all in the 2030s and 2040s. The difference between my estimate and the UN estimate is of the order of 25 to 35 million Americans by 2050, which is significant for the US, but not so significant in the context of global demographics where the numbers are much larger.
On the Demographic Dividend
The demographic dividend is an economic benefit which can occur after mortality and fertility rates decline in a given country. A decline in the mortality rate is generally followed by a decline in the fertility rate, as more women gain access to better health care and to some form of birth control. Over a period of decades, each adult and each working person will have fewer dependents to support. In the right context and with the right policies, this decline in the dependency ratio will yield a demographic dividend. The best examples of the demographic dividend are found in East Asia, and in developed countries. The demographic dividend in the US has largely been reaped and it now threatens to turn into a liability unless we enact the policies needed to deal with its aftermath. I wrote in Our Growing Inactive Population that the dependency ratio is about to reverse unless the retirement age is raised to 70 years.
In theory, Africa could be the next place to benefit from a demographic dividend. This would require not only a big decline in mortality and fertility rates, but also the adoption of government policies which foster political stability and encourage economic development. As to timing, it may be decades before the dividend appears, if at all. Outside of Africa, India could also reap a large demographic dividend.
Outside of the sheer numbers which are startling enough, there are other considerations which flow from the age distributions and economic conditions in various countries. The challenges posed by aging populations on developed economies and their government social programs are well known and documented. It is enough to say that the status quo is untenable since it would lead inevitably to an explosion in government liabilities and to a severe deterioration of the economy. This statement applies easily to North America, Europe and Japan.
Less known are the demographic issues facing developing nations. Nicholas Eberstadt of the American Enterprise Institute described them in a 2011 working paper:
"China is confronting the demographic version of "the perfect storm" and these new demographic realities may ultimately force us to revise today's received wisdom about "China's rise"."
"China's future demographic profile will differ substantially from its current population situation, mainly because of the country's low levels of fertility. Although there are some inconsistencies and problems in official Chinese population data, population specialists believe that China became a sub-replacement fertility society about two decades ago-and that birth rates have fallen far below the replacement level since then."
"In the decades immediately ahead, China will see the emergence of a growing host of essentially unmarriageable young men. This outcome will be the all but inescapable arithmetic consequence of the gender imbalance that has accompanied the country's "One Child Policy" - while ordinary human populations regularly and predictably report 103 to 105 baby boys for every 100 baby girls, China's officially reported sex ratio at birth (or SRB) was almost 120 boys for every 100 girls in 2005. This imbalance between the numbers of little boys and little girls in China sets the stage for a "marriage squeeze" of monumental proportions in the decades just ahead."
In India, Eberstadt sees a significant North-South divide correlating the birth rate with education and economic opportunity:
"[India's] dilemma can be highlighted by contrasting the prospective educational profiles of Kerala (which is now one of India's most prosperous states) and Bihar (one of its poorest). In just over a decade and a half, Kerala's working-age population will be on the brink of stagnation-but the state's working age manpower will be fairly well trained (roughly half of Keralites aged 15-64 would have high school education or better). By contrast, Bihar's working-age manpower will still be growing briskly-but as 2030 approaches, these projections suggest that well over half of working-age Biharis will have received no more than some primary schooling, and nearly a third of the state's working age manpower will have no formal education at all."
Globally, Eberstadt sees a slowdown in the growth of the working-age population:
"By the reckoning of the UN Population Division, the world's population of "working age" (conventionally, albeit somewhat imperfectly, defined as men and women 15-64 years of age) grew by 1.3 billion, or about 40%, between 1990 and 2010: a pace averaging about 1.7% a year. Given the pronounced global fall-off in fertility over the recent past, however, the world's manpower of economically-active ages is set to grow much more slowly between now and the year 2030. By the Census Bureau's projections, the absolute increase in the world's working age population for 2010-2030 would be around 900 million-400 million fewer than over the past two decades-and the projected average rate of global manpower growth for the coming decades is 0.9% per annum-that is to say, only just over half the tempo for 1990-2010."
Demographics are not the be all and end all of economics but they are one important factor among many important factors. They open or close a window of opportunity. There is always a risk in extrapolating the present to predict the future and we should view these figures with some skepticism. However the trends outlined above are undeniable, even if their magnitude turns out to be greater or smaller than the figures suggest.