Cashed Demographic Dividends: Why Inflation Could Stick

The marked acceleration in US inflation numbers this year has sparked a fierce debate about longer-term inflation expectations. At the center of the debate is the question whether high inflation is transitory, resulting from the inefficiencies caused by the sudden stop and start of economic activities, or if it will prove to be a more durable phenomenon. 

On the one hand, it seems logical that the ripple effects from the pandemic-induced lockdowns will continue to reverberate across economic markets, and that these disruptions may translate into higher prices. A good example is the shortage of electronics chips in the automotive sector, caused by cancelled orders from car manufacturers who are now queueing in the back of the line. While definitely a nuisance in the short term, it hardly seems likely that shortages of things like automotive chips or beverage cans will persist beyond the timeframe of a few years. The same thing goes for items like used cars, prices of which have been especially volatile. Market mechanisms will need some time to adjust to new conditions, but higher prices will likely work well in balancing supply and demand over the medium term. If prices subsequently stabilize, the inflation will indeed prove to be transitory. So there are some very valid reasons to believe inflation will prove to be a temporary side effect from pandemic-related supply disruptions and changes in demand.

However, a potential issue with the inflation debate is that the central question has been treated as a binary problem; is elevated inflation transitory or will it prove durable? What if it is a non-binary problem? What if inflation is elevated now because of transitory effects, but will remain elevated because of other causes? It is possible that there are both temporary inflationary effects in some products and services, and also a more fundamental change that runs underneath the surface. If that is the case, central bankers could walk into a trap. They may feel confident in seeing higher inflation as a passing phenomenon, and opt to maintain loose monetary policies at a time when deeper-running inflationary forces are gathering momentum. Inflated used car prices and beverage can shortages will likely prove temporary, but there may be a more problematic issue beneath the surface. 

Demographics and the Global Labor Force

Last year, millions of people suddenly found themselves without a job when companies hurried to relieve themselves of workers when economic activities came to a sudden stop. The economic shock that resulted from the pandemic lockdowns was the largest in recent history, but the subsequent recovery has also been quite forceful. The labor market has shown remarkable progress in clawing back the jobs lost during the lockdowns. However, the return of workers has not quite matched the speed of the economic recovery. The result can be seen in substantial shortages of workers across many sectors of the US economy. Surely there are some pandemic-related effects here as well. People may be hesitant to return to jobs with lots of human interactions out of fear of contracting covid-19, or they may have children to look after who can not go elsewhere. A third possibility is that people are still receiving unemployment benefits that were introduced during the pandemic, and may lack an economic need to return to work. These three factors can reasonably be considered temporary, and therefore the effects of labor shortages could prove transitory as well.

But there is another argument when it comes to the availability of labor. In ‘The Great Demographic Reversal’, Charles Goodheart and Manoj Pradhan argue that the world is entering a stage where labor is an increasingly scarce resource. They hypothesize that China’s accession to world markets, starting with the reforms under Deng Xiaoping in the late 1970s and accelerating after the collapse of communism in Eastern Europe, led to a dramatically more competitive global labor market. Manufacturing workers in the West increasingly found themselves in competition with low-wage workers in Asia as a result of China’s integration into the world economy. A significant number of manufacturing industries in the West were decimated as factories moved abroad or went out of business. Textiles, shoes and consumer electronics are some prominent examples of major manufacturing centers and sources of employment that experienced significant declines in the West in recent decades. A pleasant result of this low-wage labor supply is that the West came to enjoy large quantities of cheap imported goods, which acted as a significant deflationary force during the past three to four decades. 

Another and decidedly more ambiguous effect was that large numbers of factory workers in the West had to find work in other sectors or had to compete for relatively few manufacturing jobs. As globalization led markets in an increasing number of manufactured goods to become increasingly internationalized, labor markets in the West became substantially more competitive at the lower end of the education ladder. According to the authors, this resulted in significant pressure on wages at the lower end of the labor market, which resulted in increasing inequality as employees at the higher end reaped the benefits of globalization (cheap sourcing and bigger markets), while people at the lower end of the economic structure were left to deal with the negative consequences (stagnant wages → decreased living standards). Because workers in the West without a higher education increasingly found themselves without leverage to negotiate for higher wages, another major deflationary trend took hold.

The Demographic Dividend and China

According to Goodheart and Pradhan, a rapidly growing working age population supplied increasing amounts of cheap labor to world markets from the late 1970s onwards, which combined with dramatically lower additions to dependent age populations, delivered the world a so-called ‘demographic dividend’. As can be seen in the illustration below, large annual additions to the global working age population, and decreasing annual additions to the dependent age groups, led to an increasingly wide gap in annual additions to both groups. This phenomenon is called a demographic dividend because it is assumed that a large amount of working age people relative to dependent age people delivers significant economic benefits. The gap first emerged on a global scale in the late 1960s and, having narrowed considerably since its early 2000s peak, is expected to close completely in the late 2030s.

Annual changes in the amount of people added to the global working age population and the combined dependent age populations (age <15 and age >64) (WPP 2017).

The integration of massive amounts of cheap labor from (former) communist economies into world markets, combined with container shipping and information technology, led to significant deflationary pressures through the mechanisms mentioned above (cheap imports and wage competition). The result was low inflation as well as low interest rates in developed economies, with the latter also a result of the rise of major export economies like China. Uneven levels of labor costs and purchasing power between the West and developing economies led to trade inbalances, and excess dollar savings in export economies which were invested in US treasuries. 

We are currently in the early stages of those effects subsiding as China’s population is now quickly aging, and is expected shortly to go into an accelerating decline. In fact, China’s working age population already peaked in 2014 at somewhat over 1 billion people and is down by roughly 13 million since then. Thirteen million on a working age population of over 1 billion is of course a small relative decline, but the declines will rapidly accelerate in the late 2020s and 2030s, when China’s working age population is estimated to decline between 6-10 million people per year. Plus, the working age population in China will increasingly have to support rapidly increasing numbers of people of dependent age, primarily aged >65, leaving them on balance less available for work. China’s multiple decade one-child policy means it won’t be uncommon for a single grandchild to have to take care of two aging parents, as well as four grandparents. 

What I find particularly compelling about Goodheart and Pradhan’s thesis is that demographics developments are relatively predictable, at least in the short to medium term. After all, the people who will enter the labor market 20 years from now are already born. You can therefore literally count future (potential) workers today. And since it is impossible to quickly make more people when you run out of workers, the consequences of demographic changes could be quite significant. Of course, this may all seem rather far-fetched when much of what we hear is that the world has too many people rather than too few (which from an ecological standpoint is probably true). For this reason, I have put some of the data on world demographics in some graphic illustrations to visualize the issue at hand.

Population of China by major demographic age groups according to data from the UN World Population Prospects (2017). Future population numbers are based on projections using population data from the past; the farther out, the more uncertain the projections.

China’s population has aged rapidly during the 2010s and is estimated to reach its peak around 2030, while its working age population already peaked in 2014. Its demographic development is somewhat unusual because of the one-child policy implemented in the late 1970s. This policy has resulted in dramatically fewer births from the late-1970s onwards, which will lead its population to age quickly at a relatively early stage of its economic development. Fewer workers, higher prosperity and significantly more older people may lead to a situation where the amount of labor available to produce for world markets will go down substantially. Perhaps a more vivid way to stress the impact of these changes is to put the net changes in China’s working age demographic group in a separate illustration. 

The annual projected change in China’s working age population (left axis) and its annual growth rate (right axis). Data from UN WPP 2017.

As seen above, the rapid rate of growth in the working age population during the second half of the 20th century and the early 2000s has already reverted into a decline, which will quickly accelerate during the late 2020s and 2030s. To put this decline in perspective; as recently as the early 2000s China usually contributed 1 out of every 3 or 4 people added to the global working age demographic group every year. In the 2030s we will see the inverse of this incredible phenomenon, when the decline in China’s working age population will decrease the growth in global working age population by almost as much. Because China is such a large part of the world, with roughly one in five world citizens being Chinese, the potential of unlocking new supplies of labor will only work to a certain extent (India will become pivotal). In the graph below I have detailed the UN projections for the annual changes in the global working age population and its growth rate.

Annual additions to global working age demographic group (left axis) and the annual growth rate in de world working age demographic group (right axis). Data UN WPP 2017

The world witnessed substantial growth in its working age population during much of the second half of the 20th century (usually between 1% and 2.3% per annum). However, between 2003 and 2017 the annual growth rate decelerated sharply from close to 2% in 2003 to slightly below 1% per year in 2017. Between 2017-2024 growth flattens out between 0.9% and 1% per annum, before a further deceleration from the mid-2020s onwards (by 2045 it will have halved again from the 2017 rate). In the graph below I have detailed the annual net change in world working age population compared to the change in China’s working age population. The declining additions to China’s working age population detracted significantly from the annual growth in the global working age demographic group between 2003-2014. From the mid-2020s onward, larger working age declines in China will drag global net additions lower still. 

Annual change in world working age population compared to annual changes in China working age population (non-cumulative). UN WPP 2017.

Of course China will remain a significant factor in the world economy in general and many product markets specifically, but it is all but inevitable that the seemingly endless supply of cheap Chinese labor available to produce for world markets will quickly become less plentiful. Not only is it likely that China’s production will increasingly go towards satisfying domestic demand, but a growing portion of its workers will likely be needed to look after China’s own elderly. Because of its rapid aging, China’s demographic challenges will likely result in quite significant changes in its society which are all but inevitable to reverberate across the rest of the world. 

Global Aging and inflation

China’s aging population is nowhere near a unique situation; significant parts of the world are now aging quite rapidly. The group aged >65 has been the fastest growing demographic group for a long time, but will only widen its growth lead over the next 30 years. Between 2020-2050, the global group aged >65 will more than double, while the group aged <15 will expand by just 5% in total. The working age group is project to only grow by approximately 20%. What this will mean exactly for the economies of the West, China and Japan, which are aging most rapidly, is largely unclear. 

World population by major demographic age groups. The youngest demographic group has already largely flatlined and will show only modest growth going forward. The working age group and group aged >65 are still growing, but the growth rate for the working age demographic has slowed considerably. Data from the UN WPP 2017.

According to Goodheart and Pradhan, a changing balance between the dependent age groups (<15 and >65) and the working age group (15-64) in favor of the former, will lead to inflationary pressures. This is contrary to popular belief that older demographic groups tend to save more and spend less, and are therefore deflationary. The authors allege that this belief is false; the dependent age groups are less economically productive but still act as consumers of goods and services. Therefore they subtract from supply but contribute less or nothing at all to it; hence they add to inflationary pressures. The productive age group on the other hand is deflationary because they produce more goods and services than they consume. Another issue is the fact that the demographic group aged >65 is the biggest relative consumer of medical care, an expense which has shown a tendency for above-average rates of inflation.

The annual net changes in world demographic groups aged <15 and >65. Growth has largely petered out for the youngest demographic group, while the group aged >65 will continue to grow quickly. The world started adding more people aged >65 during the late 1990s and the gap has largely widened since then. Data from UN WPP 2017.

The balance between those of working age and those of dependent age (<15 and >65) can be expressed quite simply in the dependency ratio, which is a function of the sum of the groups aged <15 and >65 divided by the working age group. This ratio is displayed by the blue line in the graph below. What is notable about the world dependency ratio is that it peaked at 0.76 in the 1966-1968 period and then went into a long term decline primarily as a result of a falling global birth rate. This led to dramatically fewer <15 age people relative to working age people (mint green line). It is notable that the dependency ratio and <15 age/working age ratio move in sync from the 1950s all the way through the early 2010s, and then start to diverge dramatically as the <15 age/working age ratio continues its decline, while the dependency ratio starts to increase. This can be explained to an important degree by the aging of the baby boom generation in the West and the aging of China’s population, as evidenced by the world >65/working age ratio starting to accelerate around this time (red).

World dependency ratio, world <15/working age ratio (mint green) and the world >65/working age ratio (red). Also included are >65/working age ratios for the US, China and Japan (dotted lines).

I have included the >65/working age ratios for the US, Japan and China in part to show their relative aging to working age population dynamics. Japan (orange dot line) has already aged significantly and is currently at 0.48 people aged >65 for every person of working age (doubled from 0.24 in 1999). China (yellow dot line) is currently in the early stage of its acceleration and will increasingly go the way of Japan (although somewhat less extreme).

US Demographics and Labor Market

The shortages we are seeing in certain labor market sectors across the US right now could be an early sign of longer-lasting labor scarcity. If that is the case, the future may see a very different labor market than we have known in recent years; one where relative scarcity is the new normal. Tightness in US labor markets, if it does take hold, could become an important factor in the future direction of inflation. Let’s take a look at the major US demographic age groups below. Around the time of the last crisis, the US was adding significant amounts of people to its working age group. The working age demographic group was growing between 1 and 2 million people per year in the 2008-2012 period (a growth rate between 0.6% and 1% per year). When you are adding that many people to your working age population in the midst (and the aftermath) of a large recession, it is not surprising that unemployment takes a long time to go down. When we look at the current rate of growth in the US working age population, a completely different picture emerges. The US is adding only about 300.000 to 500.000 people per year to its working age population in the 2020-2022 period, (between 0.14% and 0.25% growth) and substantially less in the years thereafter. This is only about a quarter of the 2008-2012 growth rate.

Annual changes in the US working age demographic group (>15 but <65) against the left axis and its annual growth rate (right axis).

It seems reasonable to conclude that this will likely make for a very different labor market, especially since the working age population’s growth rate will remain very much subdued for the remainder of the 2020s. The difference between 2008 and 2020 can be explained to a significant degree by the baby boom generation, which was largely still at work in 2008 but is currently near or over the retirement age. The baby boom generation makes up a sizeable chunk of the US population and is usually defined as people born between 1946 and 1964. Since people born in 1956 reach retirement age this year, well over half of all baby boomers have already reached the retirement age. The aging of the baby boom generation causes relatively high outflows from the labor force, which likely means even a relatively slow economic recovery will absorb excess labor supply more quickly than after the 2008 crisis.

Of course, I should stress that the working age demographic is not the same thing as the labor force. The labor force is much smaller than the working age demographic because of the participation rate among other things. Another issue to consider is that in the definition used by the UNWPP the working age demographic includes people aged 15 until 64. I presume this is because it is a global metric and people in the developing world usually start work at a relatively young age. In the developed world however, people generally start work at a much later age, usually because of school and studies, although they may participate in the labor market through part-time jobs. I should also add that the labor force can adjust to a significant degree to different circumstances, while the working age demographic generally cannot (except through migration). For instance, the labor participation rate can go up when more people work past retirement age, or when more women join the work force, thus adding to labor supply. Demand for labor on the other hand can also be moderated through growth in labor productivity, usually through technological advances.

Policy Risk

A tighter labor market could have serious consequences for the supply side of the economy. Shortages resulting from interruptions in production, for example through strikes or unfilled vacancies, may lead to a more delicate supply-demand balance in a significant number of goods and services markets. Perhaps more importantly, labor may also discover its negotiating position versus employers has improved again, potentially opening up the gates for wage inflation. If this takes hold, it can spill over into the cost base of products and services, leading to higher inflationary pressures in the economy.

Increased economic sensitivity to inflation would signicantly diminish the suitability of the current loose monetary and fiscal policies. Of course, it is possible that policies will be adapted in time to reflect a change in circumstances. However, significantly tighter monetary policies may be resisted by other branches of government, because they are likely to lead to fiscal and economic pain. The consequences of more stringent monetary policy are especially unlikely to be pleasant with regards to current asset price levels.

Already there seems to be a strong political tendency in the US executive branch (and associated faction in congress) to look at the post-pandemic recovery as an opportunity to right the mistakes made in the aftermath of the Great Financial Crisis. While somewhat understandable, policy makers should realize that the post-covid recovery is taking place in very different circumstances from those of 2008. With inflationary pressures seen building across Western economies, maintaining loose monetary and fiscal policies could turn out to be a serious policy mistake. Yet some of the ideas coming out of Washington right now, where the prevailing impulse seems to be to spend ever larger amounts of borrowed money, are altogether carefree regarding inflationary risks. For the legislative and executive brances of government there is now a clear incentive to play down inflation; many Western governments, including the US, have let government debt pile up so high, inflation may well be the least painful way out. For this reason, we should expect to see central bank independence tested in the coming years.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s