Embed Notice

@LordCaramac

We are getting closer and closer to the point where it will become impossible to even simulate growth; real net growth of the economy as a whole might be over already

I don't know that this is true. One of the markers that I liked for measuring economic was the amount of labour necessary to be able to afford an hour of light. For pretty much any time before the industrial revolution, one person working for an hour could not afford to light a single room for an hour. For most of human history, you needed to work a lot longer than an hour to get an hour of lighting. Artificial light in any real quantity was the sole province of the class of people whose wealth was not tied to their labour.

When I was a child, a moderate sized room was lit by a 60-100W bulb. The cost of the bulb was largely immaterial because the cost of electricity to run it completely dominated the total cost of ownership. By the time I went to university, we were replacing those with 15-20W compact fluorescents. Now the cost of the bulbs mattered because the power cost was low, but their lifetimes were also much longer (most of mine lasted 6-10 years, whereas it was rare for me to see an incandescent last a whole year). Now, my house is mostly lit with 6W LED bulbs. The LEDs last more or less forever, the rectifiers for feeding them DC are less reliable, but the amortised cost still rounds to almost nothing.

The price cap in the UK for electricity is 24.67p per kWh. One hour of lighting consumes 0.15p of electricity. The national minimum wage for people over 21 is £12.71. So you need to work a minimum-wage job for about 0.4 seconds to be able to afford an hour of lighting.

Moving to renewable energy will probably lower some of that cost, but we're at the point where it doesn't matter.

But I don't see that across the whole of civilisation. There are a lot of things where efficiency could be improved. I think we're seeing a couple of important transitions here:

First, the twentieth century was largely dominated by big easy wins. Electrification and the internal combustion engine both had a lot of low-hanging fruit for bringing in efficiency improvements. Each efficiency improvement makes the next one harder. If you start at 1% efficient, doubling your efficiency is easy. Once you reach 51%, it's impossible (without some redefinition).

Second, and most importantly, we're in a period where the sources of productivity improvements shift. The petrochemical industry drove a load of these (both from fuel and plastics) in the twentieth century. The computing industry drove a lot of economic growth for the last 40 years because there were a lot of improvements from going from no automation to some automation, across a load of industries. Both hit diminishing returns a while ago.

This causes problems for current implementations of capitalism. The premise of capitalism is that people who are good at efficiently allocating capital are allowed to allocate more capital. Without the kind of safeguards that protect against too much concentration (very high top tax rates, wealth taxes, inheritance taxes, and so on), capital accumulates in the hands of people who were just lucky and stays there. One of the ways to be lucky is keep predicting that the industry that was successful last year will be successful next year. In times where the potential sources of growth pivot, that's a problem.

This is why we see VCs pouring money into whatever nonsense Silicon Valley 'luminaries' are spouting this week, rather than into completely different industries that would actually spur growth. And so we see shortages of capital in things like improved public transport (which would grow the economy) and an overabundance in things like buying GPUs to train bullshit generators.