By no means is this meant to be an affront to the OP or the researchers.
Part of the problem with statistics is that they can be used to prove almost any point you want to make.
I'm no doom-n-gloomer, in fact I don't care either way, but some of the stats in the OP really are awful, I'll elaborate below:
Two centuries ago, 94 per cent of people lived on $2 a day (in today's dollar value); today, only 10 per cent do.
2 centuries ago, the world population was about 1B, 94% of which would make it about 940M living on $2 in today's dollar value. 10% of today's population of ~7.5B would be ~750M. Now thats about ~200M difference, in a positive direction but it's not THAT big a difference considering the time gap. Whats also missing here is what percent is living on $3/day because if that was the other 90%, that would be bad. Also hidden in numbers can be improper methodologies, purposeful or not. In these specific numbers, they are a bit meaningless since $2 in one place is worth MUCH more than in another - and I'm not talking about exchange rates. I'm talking about cost of living. For example, the exchange rate across the U.S. is the same but a NICE 3 bedroom house usually is less than $100,000 in the country areas while it will run you a few more zeros at the end in the more expensive areas in California. So the guy making $50k/yr in the country areas is practically rich, but in Cali, he would be broke and probably sharing a residence with a roommate.
In the early 1980s, nine in 10 Chinese lived in extreme poverty; today, after more than three decades of market reforms, just one in 10 do."
Similar to above, even though the percentage change is great, the raw numbers are awful. Back in the early 80s, it was about 900,000,000 people in that category, many of whom would be dead now. Now it its about 412,000,000. The raw number is less than half the original but not as large a change as 9/10 -> 1/10 would suggest. And considering that we are talking 30+ years later - it could be better.
Unfortunately there are too many stats that are gathered/presented without giving the recipient enough information to fact check the results and make sure there is no bias.
There are lies.. Damn lies.. and Statistics!