How America’s baby bust hurts economic growth
Also: Driverless cars have an AI problem; America's spies predict America's future; the impact of alien contact on human progress
“It is impossible to make real progress in technology without gambling.”
― Freeman Dyson, Disturbing the Universe
In This Issue:
👶 What America’s baby bust means for economic growth (570 words)
🤖 Driverless cars are telling us we might have a big AI problem. Can we solve it? (621 words)
🔮 What America’s spies see in America’s future (468 words)
👽 The impact of extraterrestrial contact on human progress (579 words)
👶 What America’s baby bust means for economic growth
Americans in 2020 had the fewest babies since 1979, while the total fertility rate was the lowest since the government started tracking it in the 1930s. And demographers think there’s more going on here than just the impact of the COVID-19 pandemic. Yes, twin health and economic crises would be expected to dissuade some women from getting pregnant. But birth rates never really recovered after the Global Financial Crisis, costing the US some nearly 8 million additional people since the GFC, according to the Wall Street Journal. (See below charts.)
The typical non-pandemic explanation for those lower fertility rates: a combo of women marrying later in life, getting more education, and feeling less financially secure than their parents and grandparents. An additional explanation comes from researchers Lyman Stone and Laurie DeRose who think Americans focus too much on their jobs. As they write in The Atlantic, “When countries on the whole shift toward valuing work more, birth rates fall. … Forced to choose between the family they want and the career they want, people are opting for the latter, nudged along by policy makers hoping to encourage work.”
Assuming American preferences aren’t going to suddenly shift anytime soon, what does that mean for US economic growth? Some research finds that the older a country’s population, the lower its overall rate of entrepreneurship. From a 2014 paper: “Older workers do not possess the advantages of youth, but more significant is that when older workers occupy key positions they may block younger workers from acquiring business skills.”
Consider also the even-more fundamental growth impact from having fewer kids: Economic growth emerges from people having ideas. When lots of people have lots of good ideas, a society pushes forward the technological frontier and becomes more productive. One explanation for the post-1960s downshift in productivity growth is that ideas are becoming harder to find. So either we need more researchers or researchers need to become more productive (perhaps with the help of AI) or both.
In the 2020 paper “The End of Economic Growth? Unintended Consequences of a Declining Population,” economist Charles I. Jones makes the point that the share of the population doing research can only be so high, and thus to get more researchers you need a bigger total population. So having fewer kids may have a profound impact on idea creation (including those ideas that would deal with climate change) and economic growth, leading to a dreary “empty planet” scenario of falling population and stagnant incomes. From that paper:
In many growth models based on the discovery of new ideas, the size of the population plays a crucial role. Other things equal, a larger population means more researchers which in turn leads to more new ideas and to higher living standards. Automation could enhance our ability to produce ideas sufficiently that growth in living standards continues even with a declining population, for example. Or new discoveries could eventually reduce the mortality rate to zero, allowing the population to grow despite low fertility. Nevertheless, the emergence of negative population growth in many countries and the fact that it has profound implications for the future of economic growth make this a topic worthy of further exploration.
So maybe we’re facing a long-run race between the decline in researchers vs. improvements in how AI boosts researcher productivity. It’s also important that high productivity nations keep growing, whether through pro-baby policies or attracting global talent.
🤖 Driverless cars are telling us we might have a big AI problem. Can we solve it?
Considered without context, it was a pretty amazing thing that happened last year: Waymo opened its driverless taxi service in the suburbs southeast of Phoenix to the general public. When so many tech innovations don’t scream “You’re living in the future!” — say, social media — self-driving cars kind of do.
Now the context: Last year was supposed to be an undeniable breakthrough year for autonomous cars. Back in 2015, The Guardian offered this headline prediction: “Self-driving cars: from 2020 you will become a permanent backseat driver.” More famously, Tesla Motors CEO Elon Musk promised in 2019, “A year from now, we’ll have over a million cars with full self-driving, software … everything.”
Those are just two examples of disappointing predictions cited by the Santa Fe Institute’s Melanie Mitchell in her paper “Why AI is Harder Than We Think.” Of course, disappointing predictions seem endemic to artificial intelligence. Mitchell notes the field has “cycled several times between periods of optimistic predictions and massive investment (‘AI spring’) and periods of disappointment, loss of confidence, and reduced funding (‘AI winter’).”
So what causes these repeated forecasting errors, as well as public misunderstanding about the current state of AI? Mitchell cites four fallacies for blame: First, advances in specific tasks — such as the chess-playing Deep Blue or IBM’s question-answering Watson — are often seen as steps toward general intelligence AI. But as Mitchell quotes one engineer: “It was like claiming that the first monkey that climbed a tree was making progress towards landing on the moon.”
Second, we think easy things are easy, and hard things are hard. But it’s the opposite. Sure, AlphaGo may be the best Go player in history, but that’s something that comes relatively easy for an AI system as opposed to playing charades, Mitchell points out.
Third, AI is often described using “wishful mnemonics” — terms associated with human intelligence, such as “neural networks” and “deep learning” — that give the impression something human-like is happening. But Mitchell is skeptical: “However, such shorthand can be misleading to the public trying to understand these results (and to the media reporting on them), and can also unconsciously shape the way even AI experts think about their systems and how closely these systems resemble human intelligence.”
Finally, there’s the notion that intelligence is “all in the brain” and thus general AI can be separated from the human body. This is the (perhaps) cartoonish “ghost in the machine” or “brain in a vat” kind of AI that embraces the notion of non-corporeal “pure intelligence” that is “independent of emotions, irrationality, and constraints of the body such as the need to eat and sleep.” It’s only this constraint-free and common-sense-free version of AI that might — and Mitchell cites a famous example here — use up all the Earth’s resources to produce paper clips if it’s programmed to produce paper clips.
More on AI and the “common sense” problem — and maybe a way to solve it:
While common sense includes the vast amount of knowledge we humans have about the world, it also requires being able to use that knowledge to recognize and make predictions about the situations we encounter, and to guide our actions in those situations. … No one yet knows how to capture such knowledge or abilities in machines. This is the current frontier of AI research, and one encouraging way forward is to tap into what’s known about the development of these abilities in young children. Interestingly, this was the approach recommended by Alan Turing in his 1950 paper that introduced the Turing test. Turing asks, “Instead of trying to produce a programme to simulate the adult mind, why not rather try to produce one which simulates the child’s?”
🔮 What America’s spies see in America’s future
America's National Intelligence Council — a group of analysts who report to the Director of National Intelligence — thinks the coming decades will “bring more intense and cascading global challenges ranging from disease to climate change to the disruptions from new technologies and financial crises.” There’s a lot of that sort of predictably dreary strategic forecasting in the NIC’s quadrennial “Global Trends” report released last month.
More interesting is the report’s outline of various scenarios created by possible responses to those challenges. And most have a dystopian streak in them. In “A World Adrift,” for instance, the world of 2040 is “directionless, chaotic, and volatile” where “China is taking advantage of the West’s troubles to expand its international influence, especially in Asia.” No one is doing much about climate change. In general, the report is worried a lot about climate change, drawbridge-up economic policies, Chinese aggression, and a lack of American leadership.
At times the report reminds you of that scene in Avengers: Infinity War where Doctor Strange peers into 14,000,605 alternate futures and finds only one way for the heroes to defeat Thanos. The most optimistic scenario is “Renaissance of Democracies.” Here’s that world:
Open, democratic systems proved better able to foster scientific research and technological innovation, catalyzing an economic boom. Strong economic growth, in turn, enabled democracies to meet many domestic needs, address global challenges, and counter rivals.
The combination of better service provision and anticorruption efforts helped restore public trust in institutions and eventually mended many fractured societies. Strong differences in public preferences and beliefs remained but these were worked out democratically.
US leadership proved central to multilateral coordination and focus on global challenges, building on established alliances and international institutions. A revival in the EU and United Kingdom, spurred on by technological innovation and economic growth, was key to broader success.
Over time, the combination of severe repression, stalled economic growth, and mounting demographic pressures undermined established authoritarian regimes in China and Russia, making them less predictable and more aggressive in their neighborhoods.
So not a utopia in any way, but a world I’d take my chances in. It’s one built on technological progress, economic growth, and open societies where a culture of civil but vigorous debate dominates. Moreover, the reinvigoration of the West would persuade emerging and developing countries that liberal capitalist democracies were more adaptable, more resilient, and better able to cope with growing global challenges than authoritarian ones. Is such a scenario possible? The success of vaccine creation and distribution in much of the West suggests it might be. Indeed, the report directly cites those achievements as having the potential to focus “global attention on the importance of scientific research, innovation, and technological development to address emerging global challenges.” I guess Faster, Please! is also picking the right things to focus on.
👽 The impact of extraterrestrial contact on human progress
How should investors trade a nuclear war? I hazily recall a book where the opening vignette imagined a trader buying and selling on the floor of a commodities exchange. Suddenly there’s breaking news: The Soviet Union has launched a massive nuclear attack. Markets start to collapse. What should the trader do? Buy, buy, buy, of course. If it’s a false alarm, markets will quickly rally. And if it’s not a false alarm? Then you’re probably not going to care about your long and short positions. Bye, bye, bye.
I was reminded of that “what if” story when reading a recent Bloomberg column by economist Tyler Cowen who speculated how the discovery of alien life might affect financial markets. The prompt for the piece is the expected government report on what Washington now calls “Unidentified Aerial Phenomena.” Among Cowen’s predictions: If alien drones were buzzing around — of the sort Navy pilots have seen — the dollar would soar since it would be assumed the US government probably had the best info on the aliens. But beyond that, maybe not much reaction. Defense stocks might rise at first, but what could we do against a civilization that crossed an ocean of stars to attack us?* And if that’s as far as it went — alien craft appearing now and then or even hovering above our cities — we’d probably get used to it and go on with our lives.
Maybe, but I can also imagine a different economic scenario. When Roger Bannister broke the four-minute mile mark in May 1954, he also broke a record that had stood for nearly a decade. Then the barrier was broken again in June. More sub-4 minute miles were soon to follow. To this day, many see Bannister’s achievement as an inspirational feat of such power that it spurred humanity to shed self-imposed psychological shackles and make an evolutionary leap forward — well, at least those members of humanity who were elite runners.
A great story, but unfortunately one that doesn’t appear to be true. Over the next few years, there were only ten other sub-4 runners. And why did it take so long after getting the record down to 4:01 in 1945? “It’s much more likely that the stagnation was attributed to the war and the breakthrough was a return to sport along with the modernization of training which occurred during the 50’s and early 60’s,” running coach Steve Magness wrote in 2017.
But I wonder if the urban myth of Bannister and the psychological barrier might prove more realistic in the case of alien arrival. Their indisputable presence and technological advancement — even if we never got hold of their ships and gadgets — would show humanity that we’re nowhere close to our potential. Good ideas might be getting harder to find, but we’re nowhere close to finding all of them. Maybe there’s a warp engine out there somewhere, waiting for us.
I would like to think there would be renewed interest in science of all sorts, in discovery, in maximizing human potential — especially when we could see such vivid evidence that advanced civilizations won’t ultimately destroy themselves through war or planetary negligence. There might also be a greater sense of humanity’s shared destiny, as President Reagan noted in his 1987 UN speech where he speculated an alien attack would bring together the nations of the world. But maybe it doesn’t need to be an attack, even a friendly “hello” might do.
*Back in 2011, economist Paul Krugman impishly suggested the U.S. government deal with the slow post-financial crisis recovery by faking an alien invasion to spur a massive World War II-style defense buildup. John Maynard Keynes meets Orson Welles meets Adrian Veidt.
Elon Musk’s Golden Age of Tech Innovation Is Coming - Tae Kim, Bloomberg Opinion | Some big Roaring Twenties energy here. “For the young graduates going after a career in tech rather than pursuing a more traditional path, there’s scarcely been a better time in terms of the potential monetary reward. … According to Crunchbase, nearly $73 billion was invested in growth-stage North American startups during the March quarter, double the prior year. … With the increasing inflow of talent and large amounts of capital going after such exciting ideas, we may be on the brink of a golden age for innovation.”
Despite Chip Shortage, Chip Innovation Is Booming - New York Times | Amid concerns about scarce chip supplies and reliance on overseas production, there’s also a flood of new investment and companies. “Equity investors for years viewed semiconductor companies as too costly to set up, but in 2020 plowed more than $12 billion into 407 chip-related companies, according to CB Insights. Though a tiny fraction of all venture capital investments, that was more than double what the industry received in 2019 and eight times the total for 2016. Synopsys is tracking more than 200 start-ups designing chips for artificial intelligence, the ultrahot technology powering everything from smart speakers to self-driving cars.”
Poverty Reduction and Economic Growth - Lant Pritchett | A couple facts worth noting from this analysis. The first one, you might be familiar with: “In the last quarter century, 1.1 billion people, about one-seventh of the world’s population, have been lifted out of extreme poverty.” The second one, maybe not: “There is no instance of a country achieving a headcount poverty rate below 1/3 of its population (at moderate poverty line of $5.50) without achieving the median consumption of that of Mexico.” In other words, Pritchett concludes, lowering global poverty will require broad-based growth and raising the productivity of the typical person in a poor country, which is a key source of national income growth.
In a NASA simulation, scientists concluded they couldn't stop an asteroid from decimating Europe - Business Insider | Forget about some bolt-from-the-blue scenario. Even with six months warning, a group of American and European experts “determined that none of Earth's existing technologies could stop the asteroid from striking given the six-month time frame of the simulation.” The deflection space missions couldn’t get off the ground in time. The only feasible option was evacuation. That said, NASA still views deflection (as opposed to nuclear detonation) as the most promising strategy, given enough heads up. “Later this year, the agency is scheduled to launch a test of such a technology. The Double Asteroid Redirection Test will send a spacecraft to the asteroid Dimorphos and purposefully hit it in the fall of 2022. NASA hopes that collision will change Dimorphos' orbit.”