4 ms·
True. Another example of why that thought-terminating cliché à la '100 companies are responsible for 71% of global emissions, so my personal choices are irrelev
by AftHurrahWinch 4y ago
True. Another example of why that thought-terminating cliché à la '100 companies are responsible for 71% of global emissions, so my personal choices are irrelevant' is so silly.
Amazon's net emissions are up 18% because hundreds of millions of individuals choose to purchase from them.
The carbon-intensity metric is interesting:
> a measure that divides its emissions by gross merchandise sales
If emissions stay the same, but prices go up... say there is significant inflation, or a supply chain crunch which leads to price-gouging... then the denominator increases. Want to have a great carbon-intensity KPI? Better get that revenue up! Which is somewhat ironic.
Ultimately, Amazon's net emissions are up 18% and how efficiently they produce them isn't a factor in determining those emission's climate forcing effect. I.e., if you and I were on a boat in the ocean and I'm prying up planks from the hull to make children's toys, you would not care how efficiently I am making those toys.
- s1artibartfast 4y agoPoint taken regarding the the dollar denominated KPI opposed to goods shipped. >if you and I were on a boat in the ocean and I'm prying up planks from the hull to make children's toys, you would not care how efficiently I am making those toys. I would care about efficiency if the boat is full of other toy makers and people demanding toys. In that case, I would care very much about making sure that the toys are coming from the most efferent makers. I don't think is fair to assume that amazon business growth is entirely newly stimulated consumption. that is to say, they aren't capturing sales from other distributors and companies. It may well be the case that the carbon footprint for a amazon order is worse than driving to a brick and mortal Walmart. That would be a lot more valuable insight.