Teach the Unemployed to Code? yeah right

Bill Ramz Paul Gates has another video

You can tell he has an engineering background because his thought process is concise and direct, and he doesn’t ramble that much. I agree that teaching the employed to code as a solution to technological unemployment and globalization, is delusional. It’s hard enough teaching junior and high school students algebra or how a logarithm works. Coding is many magnitudes harder than that. I posit that if you selected 10,000 adult Americans from the general population and asked them to code a simple JavaScript to, say, execute a variable after after clicking a button and pasting something into a form, maybe only 1 or 2 would be able to do it, and this assuming unlimited time and access to online tutorials.

To use myself as an example, I got a near perfect score on the math part of the SAT in my early teens and had a solid grasp of calculus, too, so I perfectly fit the profile of someone who should be able pick up coding easily, yet not long ago ago I was trying get some elementary command line ‘thing’ to work, and eventually after 3 hours of scouring tutorials and being stuck, I I finally found a solution that solved part of the problem I had in mind. I had the general idea right but there was some problem with the directory configuration. If I had watched the online tutorials first maybe I would have found a solution faster. I remember watching a video on some physics concept and understanding it perfectly after just a single viewing even though I had never studied it before, including the math, yet I couldn’t get the command line to work. That just gives you an idea of how hard coding is and the rarity of talent involved in being competent at it. If even people who should be able to code and outwardly have the aptitude to do it, struggle at it, there is little to no hope for the general population. Even if, generously speaking, 1% of the general population is smart enough to code competently, probably only a small percent of that 1% is conscientiousness enough to persist for hours or even days if stuck.

Also, by coding I don’t mean buying a php or python book and doing the examples, but with some modifications by adjusting the variables. That is not coding. Starting with the solution and working your way to the answer is not coding, but rather coding means starting with a problem and working out a solution. I suspect that many people who think they know how to code, are in reality only leaning how to follow tutorials in an extremely artificial environment. That is better than nothing because at least you’ll have familiarized yourself with the variables and the general concept of the programming language, but it’s a far fry from being able to use said language to solve a problem that is not directly explicated in tutorial-form.

One way to increase the number of competent coders is through immersion starting at a very young age, similar to how how Brazil and other Latin American countries train soccer players. Children who have the potential to be good at coding would be selected at an early age, possibly as young as 7, to be immersed in coding, so that by the time they are in their late teens, those who remain will have presumably mastered it. Coding isn’t like cramming for a calculus test the night before the test by plugging and chugging. It’s about having a set of tools and knowing how to implement them to solve a wide range of problems, whether it’s font-end, back-end, or administrative. But of course, this would cost a lot of money.

Overall, for reasons discussed here, I’m not that concerned about technological unemployment. Low-skilled jobs such as food preparation seem to be impervious to automation. But this does not mean that new jobs will pay as well as old ones, which is what Andrew Yang is getting at. Manufacturing jobs tend to pay well relative to the skill involved, but service sector jobs pay more poorly, often lack unionization, and are more demanding. The result is a bifurcated workforce with a lot of crummy low-paying jobs in the service sector, some high-paying creative-class jobs as well as high-paying jobs that involve specialization such as legal or medical work, but insufficient jobs for people who are smart enough to advance beyond low-skilled labor but not smart enough for professional or white-collar work. Most people are just barely competent enough to do their jobs and not be fired. There is a restaurant where I order takeout, and I have to remind the the cashier to put the plastic utensils in the bag, because they kept forgetting. At another restaurant, I’ve seen an employee struggle to do something as simple as fold a Chinese takeout box.

Even after 15 years of schooling, many high school and even college graduates struggle at the very basics, such as solving a set of linear equations or a writing a 500-word essay that is at least somewhat coherent and isn’t total crap. This is partly a reflection of the failure of America’s education system, and private schools are not immune to this, yet we’re expecting these very people in less time than it takes for the unemployment benefits to run out to become master wordsmiths or to become master coders or to become master welders (even though welding doesn’t pay that much and requires tons of training and expensive certification despite all the hype). Policy makers and pundits overestimate the malleability and competence of the general public to pull themselves out of poverty or to learn new skills, or for people who are not even smart enough to fold a box to handle a high-powered blow torch.

The $1,000/month Yang is promising, although obviously prohibitively expensive, from a marginal utility standpoint would help the lowest of income earners the most, as well as the lower-middle class, who have families and could use the extra $1,000/month for childcare and other necessities. I think the UBI does more to help low-income Americans than the usual policy that politicians trot out. As discussed above, retraining the workforce seems impractical. As discussed above, people are who are middle-aged and of average intelligence are not suddenly going to attain economically competitive skills such as coding or website design. Trump’s trade and tariffs deals have not, as far I know and according to most economists, created jobs domestically for low-skilled and blue collar workers. Other candidates such as Warren or Sanders think the solution involves more education, student loan forgiveness, and cheaper education, as if people with IQs of 90-100 stand to benefit from more sociology courses and the added competition of more college graduates entering the workforce and competing for an already scarce pool of middle-skilled jobs. That’s not to say that Yang’s UBI is necessarily a good idea or optimal, but the ideas brought forth by other politicians are on the surface demonstrably bad or have been shown to not work. What we need is policy that addresses these challenges rather than paints over them with hope.