Here’s something I stumbled across just before Christmas, and I decided against including it in my last newsletter of 2020 because I didn’t the year to end on such a note.
I’m trying to educate myself about economic theories to better understand how things could work differently (and hopefully better).
Anyway. So, I’ve read about something called Non-Accelerating Inflation Rate of Unemployment, and it sounds quite intimidating, but it actually isn’t that complicated.
Let me explain. A key concept of capitalism is inflation, basically meaning that the value of money decreases over time.
Capitalism needs inflation, because it is an incentive to spend your money: If your money has less value a year from now, better spend it on consuming something, or invest it in hope for profits. There is no use of just sitting on it.
The cousin of inflation - deflation - is pretty dangerous to capitalism, because it means the value of money increases - a very effective deterrent for investments and consumption.
Non-Accelerating Inflation Rate of Unemployment - NAIRU for short - postulates not only a connection between inflation and unemployment, but also the ability to control inflation with unemployment: More unemployment means less inflation, less unemployment means more inflation.
At NAIRU, inflation is steady.
I find this unsettling - it means nothing less than our economic system needs unemployment (around 5% to 6%, at least) in order to function properly. If everybody would have work, and would have the material means to consume, inflation would skyrocket: If everyone can afford stuff, stuff will get more expensive immediately, until the ones at the “bottom” are shut out again, people can’t consume anymore and enough jobs are destroyed to get back to the NAIRU.
|
|
The Stuff we make outweigh all Life on Earth
|
|
|
People say we are living in the Anthropocene a geological epoch that started around 2.000 years ago, and is characterised by large-scale human intervention that forms and defines our planet like geological forces in the past.
Its most commonly referred to effects are climate change and the degradation of bio diversity, but another interesting dimension is the creation of things.
In 2020, the stuff humans make finally outweigh all biomass on earth. Meaning that "Roads, houses, shopping malls, fishing vessels, printer paper, coffee mugs, smartphones and all the other infrastructure of daily life now weigh in at approximately 1.1 trillion metric tons—equal to the combined dry weight of all plants, animals, fungi, bacteria, archaea and protists on the planet.” (Scientific American)
And this is only the stuff we are actively using. If we include all of the waste we produced, our artificial stuff would have surpassed biomass in 2003.
The speed of growth is amazing - in 1900, the weight of artificial things was just 3% of the weight of the biomass.
And some details are equally irritating - the total weight of plastic (8 gigatons) is double the weight of all living animals (4 gigatons).
|
|
|
Knowledge workers and common mechanisms of worker organisation like unions don’t go very well together. First, the knowledge worker is pretty far away from the industrial worker commonly associated with unions and labour organisation. Their work place looks like white collar, but everything else is resembling blue color more and more.
Second, most tech companies actively discourage labour organisation, emphasising the discrepancies between “traditional” labour and knowledge work.
It’s good to see that knowledge work is finally getting organised, and people no longer buy into the fantasy that what is good for your employer is also inherently good for you (aka “the rising tide does not float all the boats.”).
|
|
Facial Recognition Algorithm recognises your Political Party Preferences
|
|
|
A really weird thing about Machine Learning is that if you add enough input parameters, it becomes completely opaque to how your algorithm really works. And the results can be pretty surprising, and disturbing.
Researchers found that their facial recognition algorithm was not only able to detect faces it already has seen, but also predict those peoples (self-reported) political preferences.
While humans can predict another persons political preferences based on an image of their face around 55% of the time (a bit better than random guessing), the algorithm was able to correctly predict it 71% of the time.
This is disturbing, but not necessarily surprising. What is surprising, though, is that no-one can tell us what kind of factors the algorithm uses as input. Is it the bone structure, the lighting of the image, the pose, the smile? Isolating any of these factors did not reveal what the algorithm bases its decisions on.
This is the new dystopia: We are ruled by algorithms that work, but no-one knows how and why.
|
|
|
I hope this wasn’t a too depressing start into another year of Let’s be Fwends. But sometimes, you have to look the things you want to change directly in the eye. Thanks for sticking with me in 2021! 👀
|
|
|
|
|
