I reply: You wouldn’t see people working 60-hour weeks, at jobs where they have to smile and bear it when their bosses abuse them.
I appreciate the concrete, illustrative examples used in this discussion, but I also want to recognize that they are only the beginnings of a “real” answer to the question of what it would be like to not be poor.
In other words, in an attempt to describe what he sees as poverty, I think Eliezer has taken the strategy of pointing to a few points in Thingspace and saying “here are some points; the stuff over here around these points is roughly what I’m trying to gesture at”. He hasn’t taken too much of a stab at drawing the boundaries. I’d like to take a small stab at drawing some boundaries.
It seems to me that poverty is about QALYs. Let’s wave our hands a bit and say that QALYs are a function of 1) the “cards you’re dealt” and 2) how you “play your hand”. With that, I think that we can think about poverty as happening when someone is dealt cards that make it “difficult” for them to have “enough” QALYs.
This happens in our world when you have to spend 40 hours a week smiling and bearing it. It happens in Anoxan when you take shallow breaths to conserve oxygen for your kids. And it happened to hunter-gatherers in times of scarcity.
There are many circumstances that can make it difficult to live a happy life. And as Eliezer calls out, it is quite possible for one “bad apple circumstance”, like an Anoxan resident not having enough oxygen, to spoil the bunch. For you to enjoy abundance in a lot of areas but scarcity in one/few other areas, and for the scarcity to be enough to drive poverty despite the abundance. I suppose then that poverty is driven in large part by the strength of the “weakest link”.
Suppose that we have a truly “quality-adjusted” QALY measure, where time spent working “at jobs where they have to smile and bear it when their bosses abuse them” counts as zero, alongside other unpleasant but necessary tasks. We also count time spent sleeping as zero. It might be clearer to label this measure as “quality hours”.
(Maybe we count especially good times as double or triple, and this helps us understand people working hard to earn enough for a vacation or wedding or some other memorable experience)
In this model we could define absolute poverty based on the absolute number of quality hours per year. Maybe we set an arbitrary threshold at 100 quality hours per year. If a hypothetical medieval peasant is working every hour they are awake, except that their lord gives them Christmas off, they have 8 quality hours per year and are in poverty. If a poor Anoxian spends all their non-work time sleeping because of the low oxygen supply, except for an hour a week reading books with their kids, they have 52 quality hours per year and are in poverty.
This type of measurement wouldn’t have the same distorted effects of partial abundance, compared to the $/day metric that is commonly used. I think it would still show significant progress in quality hours, with extended childhood, longer retirement, and labor-saving devices. I think UBI experiments would likely continue to show improvements when measured with quality hours.
I appreciate the concrete, illustrative examples used in this discussion, but I also want to recognize that they are only the beginnings of a “real” answer to the question of what it would be like to not be poor.
In other words, in an attempt to describe what he sees as poverty, I think Eliezer has taken the strategy of pointing to a few points in Thingspace and saying “here are some points; the stuff over here around these points is roughly what I’m trying to gesture at”. He hasn’t taken too much of a stab at drawing the boundaries. I’d like to take a small stab at drawing some boundaries.
It seems to me that poverty is about QALYs. Let’s wave our hands a bit and say that QALYs are a function of 1) the “cards you’re dealt” and 2) how you “play your hand”. With that, I think that we can think about poverty as happening when someone is dealt cards that make it “difficult” for them to have “enough” QALYs.
This happens in our world when you have to spend 40 hours a week smiling and bearing it. It happens in Anoxan when you take shallow breaths to conserve oxygen for your kids. And it happened to hunter-gatherers in times of scarcity.
There are many circumstances that can make it difficult to live a happy life. And as Eliezer calls out, it is quite possible for one “bad apple circumstance”, like an Anoxan resident not having enough oxygen, to spoil the bunch. For you to enjoy abundance in a lot of areas but scarcity in one/few other areas, and for the scarcity to be enough to drive poverty despite the abundance. I suppose then that poverty is driven in large part by the strength of the “weakest link”.
Suppose that we have a truly “quality-adjusted” QALY measure, where time spent working “at jobs where they have to smile and bear it when their bosses abuse them” counts as zero, alongside other unpleasant but necessary tasks. We also count time spent sleeping as zero. It might be clearer to label this measure as “quality hours”. (Maybe we count especially good times as double or triple, and this helps us understand people working hard to earn enough for a vacation or wedding or some other memorable experience)
In this model we could define absolute poverty based on the absolute number of quality hours per year. Maybe we set an arbitrary threshold at 100 quality hours per year. If a hypothetical medieval peasant is working every hour they are awake, except that their lord gives them Christmas off, they have 8 quality hours per year and are in poverty. If a poor Anoxian spends all their non-work time sleeping because of the low oxygen supply, except for an hour a week reading books with their kids, they have 52 quality hours per year and are in poverty.
This type of measurement wouldn’t have the same distorted effects of partial abundance, compared to the $/day metric that is commonly used. I think it would still show significant progress in quality hours, with extended childhood, longer retirement, and labor-saving devices. I think UBI experiments would likely continue to show improvements when measured with quality hours.