TAICHUNG (Taiwan News) — So far in this series of columns on Taiwan’s looming demographic disaster, we have examined the unsustainability of Taiwan’s birthrate and possible solutions.
For example, the birthrate decline over 40 years has left only around 900,000 people in the 0-4 age bracket, compared to 2.2 million aged 40-44. The working-age-to-dependent ratio, which was 14.8:1 in 1980, fell to 4:1 in 2022 and is projected to be 2.7:1 in 2030, and a startling 1.1:1 in 2070.
Even if the birthrate is boosted dramatically, it is unlikely that parents will be willing to raise the five or six children necessary to restore the lost workforce. Regardless, Taiwan’s labor force will steeply decline for the next few decades.
In our previous column, we examined how immigration has been Taiwan’s go-to trend for handling the shortfall, and how it is likely to sharply surge going forward.
The other big trend that will transform the labor force and alleviate pressures is technology. Of course, Taiwanese factories have been using industrial robots, and little floor-cleaning robots have been skittering around scaring cats for some time now. The world is now on the cusp of a robot revolution that will dramatically change the economy and people’s lives.
Last year, the world was astounded at the artificial intelligence (AI) capabilities of ChatGPT, but once AI is paired with humanoid robots and released onto the market it will be an even bigger game changer. They already exist.
Humanoid robots are being trialed and trained at Amazon in the United States, and the first factory mass-producing robots will have products available next year. Vast amounts of capital are being pumped into hundreds of companies, and the progress in the last few years has been astounding.
In 2021, the ever-erratic Elon Musk was widely mocked when announcing that Tesla was entering the humanoid robotic market when he presented as his model a person in a robot suit. Shockingly, the latest Tesla Optimus Gen 2 robot has so many stunning features that many thought the promotional video was faked using CGI.
It has remarkable mobility and flexibility and has touch sensors sensitive enough to handle an egg in the promotional video. They also demonstrated it can work out tasks on its own.
It is still fairly stupid, but that will change very rapidly. Once one robot figures something out or is taught something, other robots will be able to access that learning, compounding their knowledge base in the way that ChatGPT did with language.
We live in a world designed for humans, so the form factor of a humanoid robot is important. Being able to tackle stairs, maneuver throughout a workspace, or navigate homes will all be crucial to their acceptance and utility.
The Optimus Gen 2 can lift reasonably heavy loads, move around at a reasonable pace, and even stand on one leg in a yoga pose. It can also work for 16 hours straight, while other companies are working on robots with replaceable battery packs so they can be operated 24 hours a day.
While the Tesla robot is intended for general-purpose tasks in the workplace and home, others are more specialized, like the Digit robots at Amazon. There are even robot ‘boyfriends’ and ‘girlfriends’ that can carry on conversations and, ahem, provide other sorts of comforts.
Economic implications
Some are concerned these humanoid robots will take away jobs. However, when the problem is a massive labor shortage, they could be a big help in filling the gaps.
Some tasks will still be more suited to humans for quite some time, though there could come a day when that is not the case. That will require rethinking the entire basis of how the economy works.
They should also boost economic productivity and drive down costs over the long run, making goods cheaper and more abundant. There have been discussions about shifting to a universal basic income when this becomes a reality, though there are other possibilities.
Humanoid robots will not likely be readily used in Taiwan until late in this decade or the early 2030s, as early models will likely be expensive, buggy, and in short supply. Elon Musk claims his robot will be cheaper than a Tesla car, but his past predictions on future prices have almost always been off the mark by years at the very least.
In a few years as they work out the kinks, reach economies of scale, and put the robots to work making more robots, the costs should come down, and it will be feasible to employ them at a wide scale.
However, as workers, robots do come with one major problem: They are not consumers. For that, we need people earning incomes, paying taxes, and buying stuff. Companies could be taxed on their robots, but the robots still will not likely be earning salaries to buy things on their own any time soon.
The long-term economic impact of a shrinking population on the economy combined with a rise in AI-capable robots is unprecedented and hard to predict. Just looking at the shrinking population, one might expect a shrinking economy to come with it as the number of consumers shrinks.
On the other hand, AI and robots have the potential to massively boost productivity, which is one of the best methods of boosting economic growth. Unfortunately, this could lead to a situation where the economy does continue to grow. Still, very little of that growth finds its way into the pocketbooks of average people, and income inequality spirals out of control.
Preparing for the path ahead
Before the 2000s, no one cared about income inequality in Taiwan because all boats were rising and incomes were rising fast. Since then, wage stagnation has set in, and even though salaries have started rising again in the last few years, recent rises in inflation have effectively wiped those gains out.
Those with wealth have been overinvesting in luxury properties while often leaving them empty. Their appetite for investing in property has priced most homes out of reach of the average worker, further depressing the prospect of having children.
A declining Taiwanese workforce, rising immigration, artificial intelligence, autonomous intelligent robots in the workforce, stubbornly high property prices despite a declining population, and rising income inequality are all going to have profound impacts on society and the economy. This unique combination of factors is unprecedented in human history.
There is no established playbook and therefore no easy answers. Discussion among policymakers, lawmakers, and the broader public should be initiated to try to find goals and solutions.
Hopefully, collective ideas and proposals can inform and refine thinking on these subjects in a positive feedback loop, leading to well-informed and well-thought-out policies. These discussions need to begin as soon as possible before events overtake our ability to prepare for them.
