5 ms·
What should this data actually be and how would one get it? You can't really do a fair experiment with random assignment where some companies use Amazon and som
by abeppu 4y ago
What should this data actually be and how would one get it? You can't really do a fair experiment with random assignment where some companies use Amazon and some don't and we see how their prices change over time. Because of the "Most Favored Nation" issue, you can't meaningfully compare prices of products between Amazon and other platforms. Monopolies and anti-competitive practices remove meaningful choices, which naturally blocks or hinders anyone from studying the outcomes of those removed choices.
- next_xibalba 4y ago> Because of the "Most Favored Nation" issue, you can't meaningfully compare prices of products between Amazon and other platforms. Sure you can. Amazon lists millions of SKUs. A huge proportion of those are sold elsewhere, often by sellers who do not also sell on Amazon. But more fundamentally, if data can't be had, how does the author know this is actually happening? Either it is an unfalsifiable claim or its not, and there should be data. The bottom line is, the author makes an extraordinary claim and thus the onus is on them to provide commensurate evidence.