Future workforce trends

Some countries are much more affected by a shrinking working age population than others. As you can see in Exhibit 2 below, of the major OECD countries, Japan and South Korea in Asia, and Spain and Italy in Europe are most affected. The numbers in Germany were as bad as they are in Italy, but the fact that Angela Merkel (who is one of the few world leaders who has seen the writing on the wall) has allowed more than a million refugees to settle in Germany in recent years has improved prospects in Germany somewhat.

The OECD country least affected by ageing is the US, where the workforce will continue to grow modestly (+0.27% annually between now and 2050). What we don’t know yet is whether that is good or bad news in terms of job security. Some researchers (e.g. BofAML) expect half of all jobs to be lost to robots before 2030. Although job security has never been an issue before in the context of technology, 50% of all jobs have never been at risk before.

Globally, the working age population will continue to grow at a reasonable rate (+0.67% annually between now and 2050), but that is almost exclusively due to rapid working age population growth in Africa (+2.56% annually). Outside of Africa, many countries face negative growth rates.

The relevance of negative workforce growth in the digital era

One of the issues I am most frequently confronted with is the relevance of negative workforce growth in an era of rapidly growing digitisation, which allows for the introduction of advanced robotics. Does it really matter anymore whether the workforce continues to grow or not? If you haven’t already done so, I would recommend you read our research paper from the 10th September, which I called A Future of Pedestrian Economic Growth (see here). In the paper, I pointed to the fact that, in the digital era with advanced robotics increasingly replacing humans on work floors, you can no longer assume that a declining workforce will translate into lower GDP growth, but neither can you assume the opposite.

Labour productivity will obviously improve if total output is roughly the same despite the workforce having been cut in half, but total factor productivity (TFP) may not. That depends on how efficiently we deploy the capital at our disposal, which is why I suggested in the September research paper that, in the digital era, TFP should replace labour productivity when we assess how efficient work processes are.

Regardless of how well robots can replace workers on work floors around the world, negative workforce growth will always affect GDP growth negatively. As workers retire, they will continue to spend but, in retirement, less money is typically available for discretionary spending, while robots spend nothing at all. In other words, total consumer spending, which is a major driver of GDP growth, can only drop.

Furthermore, revenues from income tax will shrink. Income tax is the biggest revenue source for most governments, and tax revenues will collapse if half the workforce is replaced by robots unless the tax model is changed. Even worse, government expenses stand to rise dramatically, as the elderly are much more expensive to service. This is a recipe for economic collapse, meaning that the entire tax system may have to be changed to make both ends meet.

In the context of Big Tech and how they have managed to grow profits explosively in the digital era, one wonders if Big Tech will be quite the same jolly investment, should they suddenly be made to pay their fair share of taxes like other corporates.

So, if you follow my thinking, annual workforce declines of the order of 0.5-1%, and even more in the countries most impacted by this trend, make it more than likely that the worst affected countries will flirt with recession quite regularly in the years to come. In that context, I suggest you take a look at Exhibit 3.

As you can see, the oldest continent is Europe with Africa being by far the youngest. As far as individual countries are concerned, if one excludes Monaco, the oldest country in the world is Japan with an average age of 47.3. Within the EU, the oldest country is Italy, and it will get progressively older. By 2050, more than one in five Italians will be over 60 and 6.4% will be over 80. That said, by 2030, Italy will no longer be the oldest country in Europe. That ‘honour’ will be passed to Greece not so many years from now.

At the other end of the spectrum, the youngest countries are all to be found in Africa. Four countries (Niger, Mali, Uganda and Angola) all average less than 16 years of age. Outside of Africa, the youngest country is Afghanistan (18.8 years).

The various factors affecting workforce trends

Workforce growth is, as I stated earlier, a key driver of GDP growth. It is largely a function of the age-wise composition of the population at large, and the key number is the growth of the 20-64 year olds (in OECD countries) or the 15-64 year olds (in EM countries). Having said that, other factors should also be taken into consideration. For example, lawmakers can impact workforce growth by making law changes that affect:

• the retirement age;
• women’s participation in the workforce; or
• migration trends.

Up to this point, my conclusions have been based solely on UN’s forward projections on the size of working age cohorts in different countries. In other words, I have implicitly assumed all other factors to remain constant but, in reality, they are not. There is therefore definitely a bit of wiggle room, but I would emphasize that the wiggle room is quite limited for reasons you will see in a moment.

Before I go there, I should also point out that UN’s forward projections are based on a simple assumption which may dramatically understate the true magnitude of the ageing problem. Let me explain: UN assumes that if, 20 years ago, women in country A on average gave birth to three children, but that the average has since dropped to two, women in country B, who give birth to three children today, will also be down to two 20 years from now.

This principle ignores rising living standards in most countries; it ignores higher educational levels for many women, and it ignores urbanisation, all of which have a powerful impact on birth rates. Take for example urbanisation. No other trend has had a bigger impact on birth rates than that. In short, in rural parts of the world, children are considered an asset (more hands to help on the farm) whereas, in cities, children turn into a liability (more mouths to feed).

Should the retirement age be changed?

When I look at older people today, I can see that being old means something very different to them than it meant to their parents and grandparents. A combination of not so much hard, physical work and better healthcare has led to the elderly being far more active these days, and I wouldn’t consider it inconceivable that many, but not all, could retire later than they (plan to) do.

Think about it the following way: years ago, when the DB pension model was first adopted, life expectancy for the newly retired was less than ten years, and pension models were constructed accordingly. One could even argue that, when the German chancellor, Otto von Bismarck, pioneered the DB pension system in the 1880s, life expectancy in many cases was negative.

These days, life expectancy, from the day people retire, is well over 20 years. No wonder so many DB pension plans are deeply underwater. Hence, if people were to retire a few years later, not only would GDP grow faster, but pension plans’ funding deficit would shrink too.

In practice, I can think of two caveats. Firstly, governments will face massive resistance if they move forward with plans like these. The prevailing view amongst workers and their unions is that this is not their problem, and governments shouldn’t touch what they have earned over a lifetime. Meanwhile, political parties are reluctant to touch the subject, as it is most certainly not a vote winner.

Secondly, even with the best of intentions, older workers are not as productive as their younger peers are. Productivity peaks when people are in their 30s. It then stays relatively flat over the next ten years or so, following which it begins to decline. Extending work life may therefore not have as big an impact on GDP growth as one would expect. On the other hand, you would expect this to gradually become less and less of an issue as physical demands on most jobs continue to decline.

Another point worth bearing in mind is that, by raising the retirement age, you can postpone the problem by a year or two, but you cannot eliminate it altogether. According to the OECD, the effective retirement age in the OECD troughed in the early 2000s at about 64 for men and 62 for women.

Since then, the average retirement age has risen but only modestly. One may wonder why it has only risen moderately, as physical demands on the average worker are far less stringent these days. In other words, at least in theory, the average worker should be able to work for quite a few years more than his (her) parents did. However, in practice, this is only happening to a very limited extent – mostly in the US.

Should more women be encouraged to join the workforce?

I sometimes come across even well-informed people who suggest that all we need to do for the ageing problem to go away is to increase women’s participation in the workforce. There is only one problem with that argument. The numbers don’t stack up.

According to UN, which provides regular updates on female-to-male participation rates in the global workforce, female participation began to gather momentum in mature countries in the 1980s, and the trend continues to this day. The countries with the highest female participation have a female-to-male participation ratio around 90 today, and UN data from recent years suggest that, when the ratio reaches that level, it begins to flatten out, i.e. one shouldn’t expect it to increase much beyond 90.

The one OECD country standing out in this context is Japan, where the female-to-male ratio is still below 70. However, a closer examination of that number reveals that it is only amongst the over 40-year olds that female participation is low. In the younger cohorts, Japanese women work as much as women in other OECD countries do.

It is therefore only fair to conclude that Japan probably won’t experience the boost in workforce growth from increased female participation that many expect. Furthermore, as more and more OECD countries approach the critical level of 90, higher female participation is a drop in the ocean when compared to the number of people who will retire from the workforce over the next few decades.

Having said all of that, reducing/eliminating the gender pay gap would probably encourage more women to (re)join the workforce. Skilled labour is harder to replace by robots, and well-educated women could possibly be attracted back to the workforce if the gender pay gap is reduced or, hopefully, entirely eliminated.

Should more migrants be allowed to settle in the OECD?

One of the less admirable aspects of human behaviour these days is the growing sense of nationalism, and with that follows rising populism – when the political elite decides to cater for the desires of all those who feel their concerns are being disregarded by the establishment.

Deutsche Bank runs a populism index which measures how many voters support populist parties in the G7 and, as you can see in Exhibit 4, the support for populist parties is very high at present, meaning that there is little political appetite for allowing more migrants in. I would actually assign a near 0% probability that immigration rules will be relaxed anywhere in the OECD anytime soon. The electorate simply won’t allow that to happen.

Ageing’s impact on inflation

As far as ageing’s impact on inflation is concerned, opinions are divided. Where common logic would suggest ageing to be disinflationary, as older consumers undoubtedly spend less than their younger peers do, some researchers argue that one should not look at this on an absolute basis but on a relative basis. Older consumers do indeed consume less on most goods and services, but they produce absolutely nothing; i.e. the supply curve is more affected by ageing than the demand curve is, which will lead to in higher inflation (they argue).

My initial reaction to that argument has always been that, whilst theoretically correct, evidence from Japan, which is at least ten years ahead of most other countries on the ageing curve, suggests otherwise. Japan is the world’s oldest country (ex. Monaco) and thus a prime example of what lies around the corner for the rest of the world. As we all know, inflation is almost non-existent in Japan, and I am yet to hear a valid argument as to why it should be any different in other countries.

Add to that the ongoing transition from human labour to advanced robotics. Suddenly, the supply curve may be largely unaffected by the large number of retirees, but the demand curve still is. For that reason, I am still very much in the disinflationary camp – at least when it comes to ageing.

Having said that, one item that could rise a lot in price in the years to come is medical care products, which older consumers are big users of. Prices on most medical care products are already rising much faster than consumer prices in general. Over the last 35 years, medical care prices in the US have risen almost twice as fast as general consumer prices (Exhibit 5).

Unless governments decide to regulate pricing on medical care products (which they may be forced to do eventually), this component will affect the overall CPI index more and more as the population ages. Today, medical care’s share of the US CPI index is about 8.5%, but that number can only rise with more elderly around.

Ageing’s impact on consumer spending

Data from the UK and the US on consumer spending by age group suggests a somewhat similar pattern. Broadly speaking, and as you can see in Exhibits 6 and 7 below, consumer spending in the US and the UK peaks at about the same time – when the consumer is between 30 and 50 years old.

One could argue that UK consumer spending appears to drop faster with age than US consumer spending does, but there is a reason for that. Most medical care expenditures count as consumer spending in the US, which is not the case in the UK (and in most other countries). As the older cohorts are big healthcare spenders, consumer spending amongst the elderly in the US will look more robust than it actually is. I wouldn’t expect ‘proper’ consumer spending amongst the elderly to be materially different in the two countries.

The Spending Wave

Harry Dent of Dent Research has developed a model that he calls The Spending Wave. It is calculated as the immigration-adjusted number of births every year, moved forward – in the case of the US – by 46 years in order to predict the likely peak in spending of the average household. In other words, when the Spending Wave peaks, in the US, the number of 46-year olds will peak.

Unfortunately, Harry Dent has not granted me permission to use any charts from his book, The Demographic Cliff, so you’ll have to use your imagination now, but Dent’s work has identified a remarkable link between the Spending Wave and inflation-adjusted equity returns. Apart from the post-GFC environment, where risk assets have been positively affected by QE, in the 70+ years between the mid-1950s and the 2007-2009 GFC, the correlation between US equity returns and the Spending Wave was consistently high.

In other words, Harry Dent suggests that the projected modest rise in the Spending Wave between now and the mid-2050s will result in US equity returns which will be much lower than anything we have enjoyed over the last 35-40 years. This obviously assumes a phasing out of QE over time.

Meanwhile, in Europe, the Spending Wave looks outright ugly, and the reason is simple. Generation X (Baby Boomers’ offspring) is not nearly as big in Europe as it is in the US. On that basis, it doesn’t appear unjust to conclude that European consumer spending is likely to be much more modest than US consumer spending for many years to come, and so may European equity returns when compared to US equity returns. Such a view is obviously long-term in nature. Over shorter periods of time, it is not unthinkable that European consumer spending could do better than US consumer spending.

Emerging market trends

For the first time ever, the old will soon begin to outnumber the young and, contrary to popular thinking, it is not a phenomenon only to be found in mature economies. Two-thirds of the world’s elderly live in emerging market countries, and the number of people aged 60 and over in those countries will rise from less than 400 million in 2015 to more than 1,200 million by 2050.

China is a noticeable example. Today, those over 65 account for 17.4% of the total population. By 2060, that number will have risen to 35.9%, and the share of those over 80 will grow from 1.8% to 9.8% over the same period. As a consequence, the old-age dependency ratio will rise from 17% at present to over 53% by 2060 (source: Statista). Only a handful of countries around the world will have more old people than China when measured as a percentage of the overall population.

In that context, I also find it noteworthy that it took France over 150 years for the average age of its population to double. By comparison, it has taken countries like China and Thailand only about 20 years.

The winners and losers

Countries like China with access to enormous amounts of relatively cheap labour, and the US with a comparatively benign demographic outlook, don’t have the same incentive to automate as quickly as an ageing Europe does. All over Europe – with only one or two exceptions – the workforce will shrink dramatically in the years to come. Could that be why China and the US are both falling behind Europe on the robotics implementation curve? See Exhibit 8.

At some point, though, countries like China and the US will realise that, if they want to compete internationally, not converting to robots is not an option. This could potentially lead to high unemployment and even to episodes of civic unrest. Countries like Italy, Spain, Germany, Japan and Korea have precisely the opposite problem. The workforce in those countries will decline significantly between now and 2050, forcing employers to switch to robots if they want to maintain current production levels.

Let’s assume BofAML is spot on when they predict that 50% of all manufacturing jobs will be lost to robots before 2030. In the US, 15.4 million people are employed in manufacturing (source: OECD). Consequently, 7.7 million workers stand to lose their job over the next decade. That’s a shade under 5% of the US workforce. Adding to that, let’s not forget that the US workforce will continue to grow.

Meanwhile, in Germany, 7.9 million workers are employed in manufacturing so, using BofAML’s forecast again, 3.95 million workers will lose their job. Having said that, because of ageing, the German workforce will, unlike the US workforce, shrink considerably in the years to come. The German civilian workforce is made up by 42 million people (source: OECD) but, according to UN, it is expected to shrink by 0.61% annually between now and 2050 (see Exhibit 2 again). In other words, by 2050, the German workforce should be down to about 35 million workers – a loss of seven million between now and then.

For the sake of simplicity, let’s assume the drop is in a straight line (which is not entirely correct), which implies that, by 2036, there will be four million fewer in the German workforce than there are today – precisely the number of jobs the researcher at BofAML expect Germany to lose to robots. The conclusion is straightforward – whereas automation could cause unemployment to rise meaningfully in some countries, a shrinking workforce will, to a significant degree, address the problem in other countries.

As far as I can see, assuming BofAML’s estimates are correct, not a single country will end up with labour shortages. All things being equal, this means that the combination of ageing and automation will most likely hold inflation and interest rates down pretty much everywhere.

The impact on financial markets

As already mentioned, the Spending Wave correlates quite closely with inflation-adjusted equity returns. The correlation is not perfect, but there appears to be quite a powerful link, which shouldn’t really come as a major surprise, given how important consumer spending is to corporate earnings growth.

History has shown that individuals are net buyers of equities until they reach the mid to late 50s. They then begin to switch from equities to bonds until they retire, following which they become net sellers of both equities and bonds. At least, that has been the pattern ever since records began. With very low interest rates on offer at present, combined with good returns on equities, the switch from equities to bonds could be a thing of the past, though.

Should late Boomers, who are in the process of planning their retirement now, find the bond yields currently on offer rather unattractive, the initial switch from equities to bonds could be very modest, but the wave of forced equity selling, when the retirees eventually need liquidity, will be more forceful. This point is reinforced by the fact that about 60% of total US household wealth – or more than US$74Tn – are in the hands of Baby Boomers.

Finally, in terms of European equities vs. US equities, given the ugly outlook for the European Spending Wave and the importance of consumer spending to overall economic activity, European equities could continue to underperform US equities for many years.

The link between age distribution and equity returns

In 2012, Robert Arnott and Denis Chaves published a study, which I believe to be the most comprehensive ever conducted on the effect on economic growth and financial market returns from changing demographics. Arnott and Chaves used 60 years of data across more than 100 countries in their study. The objective was to assess whether changes in the age composition of the population has a significant effect on capital market returns and/or on economic growth.

For the purpose of this paper, I’ve concentrated on equity returns (Exhibit 9). As you can see, the chart peaks at around 1% for the 50-54 age cohort, meaning that a 1% higher concentration of 50-54 olds would lead to an increase in annual excess equity returns of approximately 1%. Likewise, a 1% higher concentration of the 70+ age cohort would lead to a decrease in annual excess equity returns of about 2%. Given the large number of Boomers knocking on the 70+ door, that fact should not be ignored.

Furthermore, we know (as the Federal Reserve Bank provides such information quarterly) that total US household wealth was $123.5 trillion as at 30th September 2020. We also know that U.S. Baby Boomers own 60% of the nation’s wealth, so their impact on equity markets shouldn’t really come as a surprise. Considering that Boomers, who have been a tailwind for equity markets for many years, are now retiring in large numbers, that tailwind is likely to turn into a sizeable headwind in the foreseeable future.

So, yes, demographics matter to equity market returns, as the Japanese have come to realise. It is no coincidence that the great equity bull market has coincided with the Boomers going through their middle ages, when people tend to save the most.

A few final remarks

As far as the major economic powers are concerned, I would expect demographics to impact consumer spending the most in Korea and Japan. The Eurozone is a close second, with the UK coming a distant third. The least affected country is likely to be the US, but it would be a grave mistake to assume that demographics will not affect US consumer spending at all.

This has many negative and few positive implications. One of the most positive ones is the high degree of probability that inflation and interest rates will remain structurally low for many years to come (bar any cyclical upturns).

Finally, I should emphasise that none of the above is likely to be measurable from one year to the next. A trend like ageing is long-term in nature, and many other factors will affect consumer spending patterns more in the short term – just like we have seen COVID-19 doing immense damage almost everywhere in 2020.

Niels C. Jensen

11 January 2021

Supporting literature:

Empty Planet: The Shock of Global Population Decline, Darrell Bricker and John Ibbotson, 2019

The Demographic Cliff: How to Survive and Prosper During the Great Deflation Ahead, Harry Dent, 2014

Appendix A

The six megatrends and associated investment themes

Appendix B

Megatrend #2 and associated investment themes

As far as the impact on economic growth and public debt is concerned, Changing Demographics could be quite devastating, as ageing affects pretty much everything. Older people not only spend less money than their younger peers, they also spend it differently. Furthermore, the elderly are far more expensive to society at large than younger people are. According to estimates from the NHS in the UK, a man in his mid-80s costs the NHS 6-7 times more than a man in his mid-30s.

For those reasons, it is not unrealistic to expect ageing to have a significant impact on economic activity in the years to come, but ageing affects other things as well. Over the years, we have identified no less than seven investment themes, all of which are associated with Changing Demographics (Exhibit B1).

Currency Wars aka QE

The first country to experiment with quantitative easing (QE) was Japan in the early 2000s. In the West, QE was first applied in late 2008 by the US Federal Reserve Bank in response to the Global Financial Crisis. These days, QE is an active part of monetary policy in both Japan, Europe and the US, although the level of QE activity has varied significantly over time.

Governments around the world have noticed the German export success story and believe it is based on a very competitive currency (the euro) which is held back by the weaker members of the Eurozone. All other equal, QE definitely weakens the currency in question, and monetary authorities all over the world are facing the dilemma now that nearly everyone is doing QE in some sort of form these days, i.e. they are afraid of putting a stop to QE as their currency could rally and make the country in question uncompetitive. It is also worth pointing out that Germany has benefitted from a relatively weak currency, not so much because of QE, but because the behaviour of some other Eurozone members has held the euro back.

The tricky part is to identify uncorrelated investment strategies that stand to benefit from this investment theme. In principle, CTAs should be able to benefit, but their systematic approach has not worked particularly well in recent years – mostly due to low volatility in financial markets (CTAs need volatility to prosper). Consequently, I think this sub-trend is most effectively adopted at the tactical level, i.e. how currency risk should be managed.

Out-of-Control Public Deficits

The ageing of the populace at large can only result in fading GDP growth over time. The net effect will vary from country to country and will depend on to what extent robots can, and will, replace humans on the work floor, but every continent outside Africa (and Antarctica) will be negatively affected by this trend.

This poses a serious problem for governments around the world who are more indebted than ever. When debt is high, you need the economy to grow briskly to service existing debt. It is therefore quite likely that governments will resort to all sorts of tactics to improve economic activity, as it becomes obvious that the private sector on its own won’t be able to deliver the necessary growth.

Economic theory tells us that GDP = C + I + G + (X-M), but if neither C (consumer spending) nor I (corporate investments) will grow much in the years to come, governments may need to act. They can directly impact both G (government spending) and X-M (net exports).

One way for governments to boost G would be to improve the infrastructure, as that will have an immediate and positive impact on GDP growth. Furthermore, a better infrastructure will enhance productivity growth in both the public and the private sector, i.e. there is also a positive long-term effect from increased infrastructure spending.

An obvious place to start would be to improve the reliability of the electricity grid which is sub-standard – both in the US and in parts of Europe (Exhibit B2). Having said that, a very unfortunate demographic profile in virtually all developed countries is likely to limit governments’ spending powers, as the elderly would expect – and deserve – a service of a reasonable standard, and that is a very expensive proposition.

The Digital Revolution

The Digital Revolution may not be the most obvious outcome of Changing Demographics but, given how much the workforce will shrink in the years to come, many employers may soon have a problem, and the Digital Revolution will allow them to replace humans with robots – a trend that is already underway (Exhibit B3). So far, the industry that has been penetrated the most by robots is the automobile industry, but others will follow in the years to come.

The Digital Revolution has also affected non-industrial sectors. Let me give you an example of how impactful the Digital Revolution can be. In a growing number of airports around the world, you can now check in, hand over your luggage, go through security (assuming you are not carrying anything suspect) and all the way to the aircraft without having to deal with a single human being.

Demise of the Anglo-Saxon Growth Model

The Anglo-Saxon growth model is predicated on low savings and significant consumer debt. This growth model is at risk of being undermined in the years to come as a consequence of ageing of society.

In the Anglo-Saxon part of the world, consumers have been the primary driver of economic growth in recent years. As consumers’ purchasing powers fade, others will have to step in. Earlier, I mentioned how governments can boost GDP growth by upgrading the infrastructure, and that is also the obvious way to address this issue.

I should point out that most Anglo-Saxon countries actually benefit from a better than average demographic profile. I would be far more concerned, if a country like Japan subscribed to the Anglo-Saxon growth model, as that would be devastating. There are about 127 million inhabitants in Japan today. According to National Institute of Population and Social Security Research in Japan, by 2065, there will be no more than 88 million left, almost 40% of whom will be over the age of 65.

Retirement of the Baby Boomers

We know that the largest cohort in the world today is generation BB – the Baby Boomers – i.e. those born between 1946 and the early 1960s. According to Pew Research, there are about 70 million Boomers in the US alone. Given the size of this cohort, the economy will be affected quite dramatically as Boomers change their spending habits. For example, and as you can see in Exhibit B4 below, spending on various healthcare items rises as consumers age.

The Rise of Millennials

I have come across slightly different definitions of Millennials. Everybody seems to agree that a Millennial is a person who has reached adulthood in the early 21st century; most agree that the oldest Millennials were born in 1981, and almost everybody agrees that the youngest Millennials were born at some point in the 1990s, but the final birthyear of Millennials is disputed. If we say “early to mid-1990s”, most opinions are covered, though.

This generation will paint the colours of the world for many years to come. Millennials are, on average, better educated than previous generations. They are, more often than not, more compassionate than older generations; they are socially and environmentally conscious, and they are very tech-savvy – all assets that will serve them well in the years to come.

As a result of their social and environment consciousness, many Millennials are also left-leaning, politically speaking. We all know that such beliefs often change as you get older, but we have never before had a generation where those beliefs were so entrenched. It is therefore likely that many countries will take a left-turn politically in the not so distant future (Exhibit B5).

In the US, during the recent election campaign, President Trump made an issue out of labelling the Democrats the “Radical Left”. Those of us who have lived in Europe over the past few decades could only smile at that but, by American standards, there are definitely lefties in the Democratic party. Interestingly, a significant proportion of those people are Millennials.

From Globalisation to Localisation

In the context of economic affairs, globalisation refers to the increased economic inter-connectedness between countries. As countries all over the world have become ever more interconnected, globalisation is effectively the economic order of today.

Now, COVID-19 is toppling this economic order, but it would be incorrect to suggest that the pandemic is solely to blame. In fact, globalisation was already under pressure before the outbreak of COVID-19 (Exhibit B6). More and more people – particularly young people – have encouraged, or even demanded, that we should buy more local produce rather than shipping it from one end of the world to the other. As the Millennials grow older and take over most senior positions in society, this trend is likely to intensify.