At the start of a calendar year, looking ahead for the coming 12 months is considered useful. So think how much more worthwhile it might be if you could peer ahead for many years – decades even – confident your predictions would be at least sensible and, at best, spot on. The good news is it’s feasible to apply this approach both to the performance of a nation’s economy and its financial markets. The medium through which it’s done is demographics.
Granted, it’s not quite true that “demographics is destiny”, as suggested by Auguste Comte, a 19th century French philosopher and, arguably, the world’s first sociologist. But it comes close. The underlying reason is obvious enough – the human life cycle is comfortingly predictable. New humans invariably spend their first 15 to 20 years – and often longer in advanced economies – being cost centres. For the next 40 to 50 years they are productive assets. Then follows another 15 to 20 years as cost centres, though to varying degrees self-funding.
From this, it follows that to exploit fully the potential of demographics a nation must maximise its human capital in that middle productive section and especially towards the younger end where the innovators and entrepreneurs will congregate. Do that and it will have the best possible ‘demographic dividend’.
Since the industrial revolution, no nation has made the transition to affluence without having such a dividend. And it is probably no coincidence that four of the great national achievements of the industrial age were made by nations with the biggest demographic advantage at the time of their success; in chronological order, Great Britain, the US, Japan and China.
However, time passes. With grim predictability, people get old and, in aggregate, a nation’s working population gets old too. The demographic dividend gets spent. Back in 1950, Japan had the advantage of being the youngest nation in what became the developed world with a median age of 22. Fast forward 60 years and, by 2010, Japan had the world’s second-biggest economy, but its median age had doubled to 44. That made it the developed world’s oldest nation with – probably not coincidentally – its most stagnant economy.
Thus the economic rise and stagnation of countries can be neatly plotted on a regression chart. Chart 1 does this for Japan for the years 1965 to 2020. To smooth the path from economic juggernaut to declining force, each year’s data point for the change in gross domestic product (GDP) is the average of the most recent five. The smoothing effect emphasises the hard relationship between rising age and fading growth – the regression line falls more steeply and the fit of the data points to the line is tighter. Technically – and perhaps simplistically – an R-squared value of almost 0.8 (see chart) says that median age accounts for nearly 80 per cent of the factors that drive output. Of course, it’s not actually like that. Other factors, which can’t necessarily be so easily quantified, also affect a nation’s economic destiny; such as educational levels, property rights, economic freedom, national infrastructure and the state apparatus.
Japan’s recent experience of rising age and dipping performance may be the most dramatic on offer, but the pattern is generic. Apply the corresponding data for, most likely, any developed economy in the past 60 years and the chart would look similar.
Of more immediate relevance to investors, however, is the connection between demography and financial market returns. According to a paper produced in 2012 by Robert Arnott, a US investment adviser and serial writer of academic papers, there are “surprisingly powerful” links between the two. The paper concludes “stocks perform best when the roster of people aged 35-39 is particularly large and when the roster of people aged 45-64 is fast growing. Bonds follow a similar pattern with an age shift – they are best when the roster of people aged 50-69 is growing quickly”.
Why a strong representation of 35 to 39-year-olds should influence equity returns is not intuitively obvious except to suggest that successful entrepreneurs, bringing newly established companies to the market, might cluster within that group. That equities should be boosted when the demographic balloon is inflated around the 45-64 age group is more obvious since that is the period of ‘peak saving’, which tends to coincide with the time when earnings from employment also peak. Having built up retirement savings from investing in comparatively volatile equities, savers then tone down their risk exposure by shifting towards bonds. Hence bonds’ superior performance when the 50-69 age group gets inflated. Similarly, when retirement comes, and savings are progressively turned into cash, risky equities are more likely to be liquidated first. That will leave bonds – in simple terms, the pedestrian yet predictable asset – hanging around in diminishing portfolios for longer.
One problem with this interpretation is that it ‘jobs backwards’. It presumes that the experience of saving and investing in the 70 years since about 1950 will be the norm; that a demographic dividend will always be created in poor countries by the twin virtues of a fast-growing adult population and a falling birth rate; that the dividend will always create wealth and that a slice of the new wealth will be saved and so on.
Yet what worked so wonderfully well for Great Britain 200 years ago and to China just 20 years ago may be seeing its final iteration in the economy of Vietnam. Put simply, a big and growing pool of cheap labour operating within a comparatively stable nation state may no longer provide the pathway to affluence. At least, it is a fashionable notion among economists that such a labourforce no longer has a comparative cost advantage over developed economies. Increasingly, so the idea goes, developing countries can’t compete with the capital intensity and the need for skilled labour that go into making the high-value goods that consume a growing share of the developed world’s spending. True, developing nations can still fall back on sewing jeans, making trainers and assembling cheap electrical goods, but such bog-standard work can no longer fuel the journey to affluence.
This fear may be just another variation on the so-called Luddite fallacy that technological unemployment will lead to structural unemployment. It has been around since the wheel was invented and hasn’t yet come to pass. Even so, it may offer a useful message to investors who want to scan worldwide.
To explain, see Chart 2, which juxtaposes the median age of 32 countries selected fairly randomly with their annual growth in GDP for the 10 years 2010-20. More or less, the chart shows what would be expected – that the older the country, the lower its 10-year growth rate and vice versa. The outliers are interesting. South Africa and Brazil, two of the five so-called ‘Brics’, the biggest of the developing nations, are conspicuous underperformers.
Will that gap be made up in the coming investment generation that takes these middle-aged nations to old age? Put it this way, it stands a better chance than China, the other Brics outlier on the chart, continuing to be the major overachiever. Arguably, over the coming generation demographics will be China’s greatest threat – not just to its progress, but to its stability. Because of Mao Zedong’s infamous one-child policy its demographic dividend was especially bountiful. But that dividend will morph into a heavy burden in the next 30 years as its working population shrinks. Currently, China’s median age – 38.7 – is 5.5 years younger than western Europe’s. Thirty years hence – just one investing generation away – it will have risen to be 48, slightly older than western Europe’s. The inevitability of demography lends a chilling note to the cliché that China will get old before it gets rich.
Equally worrying may be the underperformance of Africa, encapsulated in the chart by its most populous country, Nigeria. Most of sub-Saharan Africa, where the median age is in the late teens, is still too young to have a demographic dividend. That means it may be off investors’ radar for perhaps 30 years. But it is quite feasible it will never make it onto the radar. Meanwhile, if I were to pick just one of those 32 countries as an investment on a 30-year view starting now it would probably be Vietnam, perhaps the last of the developing nations to capture a demographic dividend. That is said more as contention than assertion, but it may be worth examining sometime soon.
