The hard questions about automation
Tomorrow is the second group of the U.S. Democratic presidential primary debates and you can bet I’ll be sneakily watching it from my desk in the morning (Taipei time). Not because I am interested in seeing who will lose to Donald Trump (trust me on this, I’m from Ohio), but because of this man:
That’s Andrew Yang, the Taiwanese-American who will be standing on the debate stage next to establishment front-runners like Joe Biden and Bernie Sanders. The one-time lawyer out of Columbia Law School founded Venture for America, a fellowship program which takes top college graduates and sends them to startups in economically challenged cities across the U.S., providing mentorship and high-impact work experience along the way.
The field of 24 Democratic candidates is a crowded one, and most of them have been repeating the same old tired lines that just don’t resonate with the blue collar workers that delivered the presidency to Trump in 2016. Except Yang.
Yang’s central campaign message is that automation is destroying and will destroy millions of livelihoods as jobs are automated in this new age of AI and robotics. He likes to mention the top five job sectors in the U.S., which comprise about half of all American jobs:
- Administrative and clerical work (including call centers)
- Retail and sales
- Food service and food prep
- Trucking, driving, and transportation
- Manufacturing
It’s not difficult to recognize the magnitude of the job loss on this list that will occur over the next ten years. In fact, much of it has already started. Pizza startup Zume has raised $450M to automate pizza-making, baking, and delivery. They’ll even be able to predict pizza demand so eventually your pizza will be put in the oven before you even order it. Did I mention the ovens were located inside the (eventually driverless) delivery vans? No humans will be needed.
People will say that jobs have been eliminated all throughout history and we have turned out fine. New jobs that don’t exist now will be created. While this is true, Yang will try to convince the nation tomorrow that there will not be nearly enough new jobs created in the next 10 years to cover all the jobs that will be automated away. Millions will be newly unemployed. Life expectancy in the U.S. will continue to fall, as drug overdoses and suicides continue to increase.
To address this, Yang is proposing a version of universal basic income, an idea supported by many including Elon Musk, Mark Zuckerberg, and Marc Andreessen, in which all adult Americans would get $1,000 a month as basic income to cover basic necessities.
That’s a message that can resonate with blue collar workers, and it’s starting to show: Yang is now in third place among Democrats on prediction market PredictIt.
I’ve looked into the specifics of the plan and even had a chance to participate in a discussion with Yang when he visited my class in March. Purely from a numbers standpoint, implementing Yang’s $1,000 per month plan is doable through increased taxes and a spending reduction on welfare, prisons, healthcare, and other things that get cheaper when poor people have money. But that got me thinking: what about less developed economies like Southeast Asia?
Earlier this year for a school project, I had the opportunity to help with the series A pitch for a startup in the Philippines that is developing robots to automate away most human tasks on chicken farms. Currently, humans have to walk around the chicken barn to get the birds to move, check temperature and air quality, check on feed and water levels, and remove dead birds. As we were walking around one farm, I could barely look the farm staff in the eyes. I knew their jobs were going to be gone and I was going to help make it happen.
Automation is going to affect developing countries in a much bigger way than in the U.S. due to the nature of their economies. Take a country like Vietnam, where fully two thirds of the people work in agriculture or industry, two industries where automation will have some of the biggest impact, according to McKinsey. Automation doesn’t necessarily have to happen in a country to affect it, either — even if robots don’t take over Vietnam, another country with robots could eventually produce the same things cheaper and make Vietnamese goods uncompetitive.
Developing countries need a plan. And it needs to be more concrete and practical than “learn to code” or “transform your economy”. It’s not encouraging for them that even the U.S., the most dominant economy in the world, will struggle to deal with automation in the near future. At least the U.S. and other developed countries have strong service economies and have the means to set up safety nets and talk about giving everyone a basic income.
There is not an easy answer to this question. Mass automation is coming, and while it will greatly improve global productivity and lead to new types of jobs, countries that have not yet made the jump to a services-based economy will inevitably be left behind. I’ll definitely be following this story in ten years, while eating my robot-made pizza paid for by my $1,000 a month check.
Watch Andrew Yang take the debate stage to talk automation and AI on June 27 at 9PM EST.
