Also highly relevant:

• Steven Pinker's reply to Hickel, posted on Jerry Coyne's "Why Evolution is True" blog.
• Hickel's Guardian article suggests that the improving-world narrative is specifically being pushed by the very rich and their acolytes, but there's a much broader consensus around it than that.
• Poverty has reduced no matter where you set the threshold: the whole distribution has changed. (Same as Wiseman's first point.)
• Hickel's Guardian article suggests that it's pretty much just China that's got much less poor, but actually lots of other poor countries have improved a lot.
• Hickel's suggestion that in pre-colonial times people in those very poor countries were less poor than GDP-based measures suggest because they had highly-non-financial assets like (communal) access to water, livestock, grazing land, etc. This is "a romantic fairy tale".
• The decline in poverty shows up in metrics like life expectancy, literacy, etc., as well as when you just measure money. (Same as Wiseman's second point.)
• If you ask anyone who's actually spent time in these countries, they'll agree that they've got much better.
• Hickel is a "far leftist" and he only says this stuff because he doesn't want to face the reality that capitalism has made poor people's lives much better. (Boo to the Left!)
• Hickel's subsequent reply to Pinker, posted on his own blog.
• Much elaboration on the alleged deficiencies of number purporting to give long-term trends in poverty: they involve combining further-past figures that are mostly just about GDP (and in particular pretty much completely ignore those non-financial assets) with newer figures that actually look at consumption but are only available from 1981 onwards. You can't just splice these together and assume you get something meaningful.
• No, your account is the fairy tale without citations. (A couple of paragraphs later, he does name some books that he says support his account of things.)
• Reiteration of his argument from the Guardian article that in the colonial period the people of many poor countries got severely screwed over in ways that don't show up well in GDP graphs.
• No, he's not saying that Industrialization Is Bad, he's saying that the specific way it happened in the "global south" was bad in ways that are obscured by the figures Pinker likes to quote.
• Much elaboration on the business of whether $1.90 is a reasonable line to look at. Hickel says: no, it's way too low, for lots of reasons, and if you try to measure what's happening to poverty by looking at what's happened to the fraction of people living on <=$1.90/day then you will badly mislead yourself.
• Yes, the whole distribution has shifted, but different bits of it have shifted in very different ways; e.g., if you use $7.40/day as your threshold instead of $1.90 (a figure one person Hickel quotes has suggested) then instead of the huge decrease Pinker likes to cite you see a decrease from 71% to 58%, which doesn't seem nearly so impressive.
• Absolute numbers of people in poverty have actually increased, if you use the higher thresholds, because those reductions in proportion are cancelled out by increases in population. So the world has more very poor people now than it used to have. Especially if you ignore China, which is (1) rather a special case and (2) not where you want to be pointing if your argument is that neoliberal capitalism has made everyone better off, since China has taken a not-at-all-neoliberal path to greater prosperity. In poor countries outside East Asia, where in many cases neoliberal policies were enforced by the IMF and the World Bank, things look very different -- and in fact even relative poverty increased (if you use that $7.40 threshold) outside China between 1980 and 2000 when this was happening.
• Since that time, the biggest gains have been (1) in China and (2) in Latin American countries with socialist or social-democratic governments. (Yay to the Left! Boo to the Right!)
• Actually, we shouldn't be looking either at absolute poverty numbers or at the fraction of the world that's poor: we should be comparing the amount of poverty with the world's capacity to reduce it. That's actually worse than it used to be since the world as a whole has been getting richer much faster than the poorest parts of the world have, and it will take a shamefully long time to eradicate even "$1.90 poverty" at our present rate.
• On those other measures of quality-of-life: Yes, they're getting better, and that's good, but they don't indicate that people are getting out of poverty, and in many cases what has improved them is not the march of neoliberal capitalist globalization but simple public health interventions. And on one specific important one, hunger, Pinker's relying on figures that are bad in the same sort of way as "$1.90 poverty" figures are bad.

I'm left somewhat less than satisfied by all of this; it seems like Pinker and Hickel are ignoring some of each other's points, presumably because they don't have nice snappy responses to them.

[EDITED substantially after posting, to include rough summaries of the two things I'm citing.]

World population is roughly 2x what it was in 1970, and life expectancy is significantly up, such that there's probably 3-4x the number of anticipated future life-minutes of currently-living humans in 2019 than there was in 1970. In some views, this dwarfs any small change (positive OR negative) in the quality of those life-minutes.

The biggest benefit is going from not-living to living. Income levels are trivial in comparison.

There probably _is_ a point at which the repugnant conclusion becomes binding, and that incremental new lives aren't worth the loss of quality to existing lives. But it's many orders of magnitude out.