| Respectfully, I think this misses the point a bit. Roser himself says that there could very well be valid criticisms that Hickel is correct on and he is incorrect on. The criticism that Hickel recently raised and that you mention here (regarding the validity of pre-1981 data) sounds like it could be one of these. But productive dialogue between them was largely precluded by perceptions of bad faith and aggressive language ("mansplaining", etc) going back to before Allen's paper. Now, regarding the specific criticism about pre-1981 data, I found this[1] and this[2] which seems to claim that the older data is adjusted for non-market income. Although, I haven't digged into it properly, so I can't comment on who is correct here. "we want to emphasize is that those estimates of poverty do take into account non-market transactions such as subsistence farming.
[1] https://ourworldindata.org/poverty-home-production-and-consu... "Yes, over the last two hundred years, there has been a major shift from people farming for their own consumption towards people working for a wage and purchasing goods in the market. But historians know about history and where non-market sources of income make up a substantial part of total income, it is very obvious that money would represent a rather silly indicator of welfare.
Just as we need to adjust for price inflation, accounting for non-market sources of income is an essential part of making meaningful welfare comparisons over time. Estimates of poverty and prosperity account for both market and non-market sources of income, including the value of food grown for own consumption or other goods and services that enriched the lives of households without being sold in a market."
[2] https://ourworldindata.org/extreme-history-methods |