Resolution update: the 10% chance was fulfilled on the first day of rolls with the resolution going to Less than one million with a 3% margin over Zero and Over ten million. This may be the correct answer but no further information than my own 675,000 number was elicited in the comments
I think I should have asked how many deaths I would believe have happened after reading comments making the case for one answer or another which would increase the salience of individual efforts to find information for something that a clear resolution is not possible for. A bounty for answers that I found credible might have worked better than that
I happened to read Matt Ridley's book The Evolution of Everything a week ago and he wrote that hundreds of thousands are starving per year which I may send him an email inquiring about
.............
Resolution edit: to make last minute swings risky the answer will resolve to the market's choice with a 10% chance on day 20, stepping up by 10% on each subsequent day
First chance of resolution on Feb 10th
Net or second order effects apply. An example could be that there were arguments made that increased scale of corn production would lower unit costs
The FAO real food price index went from 73 in 2006 to an average of 110 from 2008 to 2023
1,000🏅 Top traders
| # | Name | Total profit |
|---|---|---|
| 1 | Ṁ120 | |
| 2 | Ṁ52 | |
| 3 | Ṁ47 | |
| 4 | Ṁ35 | |
| 5 | Ṁ23 |
People are also trading
@ProjectVictory I decided to resolve it to the market's choice, hoping it would make a clear piece of information if I trusted the outcome and wanted to share it on social media or anywhere. I realize this makes it possible that someone bids one answer up near the end
Do you think I should put in the description that I will resolve at 20 days with a certain probability, increasing each day after that, until the "dice roll" says 'resolve today', or the 30 days is reached?
I'm struggling painfully to find any numbers to work off of. It seems it was originally thought that 85% of the '08 spike was biofuels. Around 2012 Wood Mackenzie and The World Bank did literature reviews. Direct correlation found a weak link but models found about 25%
Around 2010 I recall reading a paper, that I thought was in The Lancet, which estimated that hunger deaths due to biofuels were from 10-20,000 to 200,000 at that time. Not knowing what the lower attribution level of the study depended on, it seems reasonable to think it was encompassing direct correlation studies under 25% so the 25% level might be mid tens of thousands
As I recall the number was annual or for one year but straightforward multiplication isn't good because poverty levels have dropped dramatically, as they have since 1990. It looks like extreme poverty has reduced from 20% of people in 2008 to 10% in 2017. So divide by two I think. It depends a lot on what number you pick and how much you trust the model number. I find it seems to make sense to pick the model number given it's a multiple lower than original estimates but I think that's a little motivated by my priors
If I adjust 200,000 at 85% attribution down to 25% I get about 60,000 annually. Divided by two, times 15 years I then get 450,000 hunger deaths
@Canucklug Actually the poverty level drop should change the baseline by .75 instead of .5 or divide by 2. This gives me 675,000
I felt like I was erring on the high side by assigning the remembered 200,000 high case deaths for an 85% price rise effect, but I realize that number could be based on the even higher estimates that resulted in an average of 85% effect attribution. So I feel like my 25% attribution choice is a little more solid than I thought