3 ms·
Is capital that is used to "produce" more capital actually productive, though? If you do finance games to instantiate money to open, say, a mine on credit and p
by grues-dinner 2y ago
Is capital that is used to "produce" more capital actually productive, though? If you do finance games to instantiate money to open, say, a mine on credit and produce iron, clearly yes by most standards. If that money is instantiated just so it can sit in a ledger and make more money to also sit in a ledger, but most of that money never actually gets used for anything "real", then maybe not.
If the concept of "production" is simply equivalent to numbers in ledgers going up, then of course capital is productive. It's almost tautologically defined, you might as well call money "productivium".
However, if you define "productivity" as something else like "building things" or "improving life quality", or "increasing biodiversity" (yes, those things can be in tension, like the iron mine, and you can wish for "bad" things too: killing people is "production" in war, perhaps) then it's not a foregone conclusion that all capital is inherently productive.
"Ah yes", you say, "but that money that's reinvested elsewhere will eventually be used for something productive in the second sense". Perhaps, perhaps not, or maybe just some, but even if so, it's still possible for someone who doesn't live in "elsewhere" to have the sense that the capital has not been productive of practical value. And many, not all, but many, people might reasonably be wondering if there's so, so much much-vaunted "value" sloshing around, why it doesn't seem to manifest in their lives.
- klabb3 2y agoYeah that’s what I meant. Capital can be used for non-zero-sum games and increase the size of the pie. That’s certainly the narrative of the current central-bank-inflationary-fiat hegemony – some fisherman who wants to buy a boat, or a farmer buying land and equipment. That’s why we (allegedly) accept the that the money in our wallets and bank accounts lose their purchasing power. However, my understanding is that the reality is quite different. That the majority of money creation (70%+) is issued as debt for asset transfers. In either case, the idea that capital, in general, is not just a proxy for, but equivalent to “productivity”, means that the system is self-reinforcing; the only outcome is success, and the only fix is more of the same system. If I argue that say functional programming is the best way to program, I have to define an external metric in which another paradigm can win. Otherwise it’s just blind belief.